Actually with Terran at +5% and Protoss at −9% we see that Zerg must be at +4%, so it's by a hair.
Here's the current top 10:
Life 1921
Polt 1837
Bomber 1818
ForGG 1816
PartinG 1798
HyuN 1768
YoDa 1751
viOLet 1738
DongRaeGu 1736
Leenock 1735
Life celebrates his seventh period in the number 1 spot, with a bigger rating gap than ever. Parting is still the only one in the top ten defending the Protoss pride. His closest compadres are Yonghwa at 1681 and Rain at 1675.
But if Terran has taken over in the Korean world, among the foreigners, it's still all Zerg.
LucifroN 1710
Sen 1677
VortiX 1669
Stephano 1638
Scarlett 1625
Nerchio 1617
Snute 1604
Happy 1558
Welmu 1555
Bunny 1526
Special mention must go to Feast, who rakes in a massive 171 points these two weeks, coming off of the Francophone Championship, ATC and two WTC EU qualifiers.
On April 18 2013 19:48 Callynn wrote: Funny how the top 1 foreigner has less points than the lowest korean. I don't understand the math model, but I don't try to understand it either
Looks like it's reflecting skill? Can someone explain how it's calculated in plain english?
it's a statistical analysis of past games. I think it correlates with skill, but it's not a one on one relation.
I don't have much knowledge of the system, but i think this at least should be true.
It's not a bad rating system and the predictions are most of the times pretty accurate but it certainly has flaws when you see ForGG as the 4th best Korean right now.
The math seems pretty solid behind it. For those people wondering why the top 10 is rather strange, the website is based on a dataset of thousands of replays. Unfortuantely pro's dont release their replays (except streamers such as ForGG) and are therefore not rated correctly
On April 18 2013 21:20 ScrubS wrote: The math seems pretty solid behind it. For those people wondering why the top 10 is rather strange, the website is based on a dataset of thousands of replays. Unfortuantely pro's dont release their replays (except streamers such as ForGG) and are therefore not rated correctly
Not so much replays, just official matches I would think.
On April 18 2013 21:23 ZenithM wrote: I like how I have no idea of the current skill level of some of the names of that list. Like, Lucifron, Sen and forGG what? :D
Edit: And no Kespa players
Haha I was just about to post the exact same thing. This list is pretty surprising.
On April 18 2013 21:20 ScrubS wrote: The math seems pretty solid behind it. For those people wondering why the top 10 is rather strange, the website is based on a dataset of thousands of replays. Unfortuantely pro's dont release their replays (except streamers such as ForGG) and are therefore not rated correctly
Not so much replays, just official matches I would think.
Even then the same problem pops up. If replays from only official matches would be taken, players from EU wouldn't be ever (or almost never) matched vs koreans, meaning that their relative ELO actually cannot be determined. This will ofcourse lead into flaws in the rankings.
The best would be if only matches/replays from the same tournament (or same class tournaments) would be considered, or at least from the same region. For example, the zotac cup and go4sc2cup are pretty similar level, and replays of those tournaments could be merged into a big dataset. GSL and GSTL goes the same, but not GSL and the zotac cup.
On April 18 2013 21:23 ZenithM wrote: I like how I have no idea of the current skill level of some of the names of that list. Like, Lucifron, Sen and forGG what? :D
Edit: And no Kespa players
Kespa players will start raising as they play more GSL (MLGs too or is that completely over?). About Sen, forGG and lucifron, like kespa players, you should take into account that they play mostly in foreign competitions so while they are good foreigner players their aligulac rating is only accurate if you compare them with people who play in those competitions aswell.
On April 18 2013 22:01 SpaceYeti wrote: IS this still factoring in performance from WoL? or was it reset with HotS? Would be interesting to see a side by side comparison.
I guess what's happening with kespa is when they all started playing they had no rating. then they only played each other so their rating didnt really go up. they only way their rating increases quickly is by beating players with established ratings, so the one that qualified for GSL are moving up. but then if they lost a game in proleague they lose to players with low rating because those players havent established a rating yet. it's kind of just a cycle that has to work itself out eventually. that's my take on it at least. could be completely wrong
List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
This seems like it's lacking a lot of data to be reliable. Also guessing WCS EU is screwing over NA players a lot, which afaik haven't really had any real tourneys in HotS? Atleast the foreigner list is far from what I'd consider the top 10, and like I said noticeably many EU players.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
I'm sure the person you are quoting isn't doubting the math behind the list, but rather that the dataset has any resemblance with the real skill level of players, when WoL stats and tourneys have only be held to specific portions of the world.
On April 18 2013 22:30 Zarahtra wrote: This seems like it's lacking a lot of data to be reliable. Also guessing WCS EU is screwing over NA players a lot, which afaik haven't really had any real tourneys in HotS? Atleast the foreigner list is far from what I'd consider the top 10, and like I said noticeably many EU players.
If I'm not mistaken, there's significantly more games in the Aligulac database than in TLPD.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
Oh, so he only wins against euros and NA players. Yeah shows how lacking the "foreigner top 10" is.
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
Oh, so he only wins against euros and NA players. Yeah shows how lacking the "foreigner top 10" is.
*he beats foreigners! I can't believe he's ranked higher than foreigners!*
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
Oh, so he only wins against euros and NA players. Yeah shows how lacking the "foreigner top 10" is.
Since there are SO many wrong interpretations in this thread already, I will try to explain some things about the site. First off, the rating is based SOLELY on results, not playstyle, not total domination, not tournament results, and not on stuff like streams/replays as some guy thought. ONLY WINS MATTER. Meaning that a win in an online qualifier Bo3 is the same as beating someone in Ro4 GSL. However, the likelihood of facing a highly rated opponent in GSL is bigger than in WCS EU Qualifier Ro64.
Here are my POV on some claims made: ForGG is not 4th best in world: Agreed Kespa players are underrated: Agreed
Picture the entire pool of points as one big circle
The circle only grows in one way, when new players enter the pool, this creates a bit of inflation (similar to chess ELO and amount of chess GMs).
How does a player gain points? Simply by overperforming, meaning that if you are set to lose 2-1 against someone, and beat him 0-2, you gain points in the given matchup (vT, vZ or vP).
Why are the Kespa players so underrated? The Kespa players have very few games compared to the rest, and they mostly play each other, meaning that the total "pool of points" that most Kespa players are playing for grows very slowly. Players like Flash, Rain, Innovation (Bogus), PartinG etc. are bringing points from the other "pools" into the Kespa pool, thus raising the amount of points being fought over.
Why is ForGG so highly rated? It is true that ForGG recently plowed a hell of a lot of mediocre terrans, boosting his TvT to a somewhat suspicious high, the amount of games ForGG plays, plus the fact that he owned A LOT of 900-1200 rated players in the specific match up gave him a lot of points, which he is losing to higher skilled but still lower rated players.
The system had a huge shock with the entry of 100+ Kespa pros, who all started at 1000 rating, but who are arguably a lot stronger. It will take a significant period of time, and amount of games from these pros, to give the system time to "transfer" points to the Kespa players, and thus accurately reflect the skill level of the Kespa players in the ranking.
Basically, the best way for this system to be 100% accurate is when everybody plays everybody, so points transfer freely, however there is still a huge disconnection between local and global scenes, as shown by TheBB in this thread. http://www.teamliquid.net/forum/viewmessage.php?topic_id=396566
On April 18 2013 22:26 Scarecrow wrote: Having only one player from SPL (currently has over half of code S) in your top 10 list makes your rating system garbage.
On April 18 2013 22:01 SpaceYeti wrote: IS this still factoring in performance from WoL? or was it reset with HotS? Would be interesting to see a side by side comparison.
The ratings make no difference between WoL and HotS. If you want to see what a HotS only rating would look like, have a look at the current state of TLPD's HotS ranking: http://www.teamliquid.net/tlpd/hots/players. Aligulac's HotS-only ranking would look pretty much the same, I would imagine. As you can see, it's not exactly useful, since there haven't been nearly enough games being played so far.
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
Oh, so he only wins against euros and NA players. Yeah shows how lacking the "foreigner top 10" is.
Other than LucifroN and MaNa I don't know who you expect to see ranked higher than him based on *recent* (and I can't stress this enough) results.
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
Oh, so he only wins against euros and NA players. Yeah shows how lacking the "foreigner top 10" is.
*he beats foreigners! I can't believe he's ranked higher than foreigners!*
Uh some of those other players not in this top 10 actually beat Koreans..just saying Then again, HotS has made it so that Koreans will dominate even harder, so if these are recent results I'm not that surprised.
Not posting any explanation of where those numbers are from, seems like a cheap advertisement for your site rather then paying TL to advertise. Not trying to be a jerk, just saying...
On April 18 2013 22:21 TAMinator wrote: List is pretty invalid if there's no KESPA players.
Because subjectivity > math?
U wot m8.
On April 18 2013 22:27 playa wrote: List is depressing. The 1 toss in the top 10 for foreigners is someone I have to ask myself if I've heard of him before, and a player whose only mu above 50% is p vs p. The stats better change a lot in HotS, or they may as well start holding Toss only tournaments.
I can see where you're coming from, but also it's your own problem if you don't follow the scene well enough to know Welmu. He's been around for a while and has always been a consistent, solid Protoss.
Lol has welmu even won anything major, or is he just another one of those guys who gets knocked out in the Ro8 of a euro/NA tournament and gets called "good"
I could say the same thing about a bunch of foreigners (who are actually far overrated compared to their actual results). And it depends what you mean by major. He won Assembly this year and placed 2nd in Copenhagen Games last year, both of which are very reputable scandinavian lans. But obviously you're going to decry that that's not good enough.
But I've watched him play and seen some pretty sick stuff.
Oh, so he only wins against euros and NA players. Yeah shows how lacking the "foreigner top 10" is.
*he beats foreigners! I can't believe he's ranked higher than foreigners!*
Uh some of those other players not in this top 10 actually beat Koreans..just saying
I expect to see Bunny rise further this year and for lucifron to assert himself as a top foreigner even more so than last year. The two seem to be in a class of their own on the EU ladder (discouting the likes of fOrGG) winrate/MMR wise (Bunny has 2 accounts with incredible MMR)
On April 19 2013 00:09 Nerski wrote: Not posting any explanation of where those numbers are from, seems like a cheap advertisement for your site rather then paying TL to advertise. Not trying to be a jerk, just saying...
I guess we just thought that by now, most posters know what Aligulac is.
The site was developed mainly for usage on this site, as an alternative to TLPD. Also advertising our site, you mean our non-profit, 0 ad site which TheBB pays for out of his own pocket?
I'm sorry but I see no reason to go to your website what so ever.
Around 20k more games from EU/NA tournaments primarily. A monthly balance report. Predictive capabilities. Hierarchial access to matches from lots of events. (Probably every decent event worth mentioning). 0 ads. Earnings of players.
The sample size of this data is too small so the rankings are too far from current realities to be posting inflammatory "top ten lists" or claims that a race's dominance is over (although coincidentally it looks like that may be the case as tvz is looking reasonable at the highest level). The statistics and math is very interesting to a psyche major but also a bit disappointing that you haven't included more commmentary with this post, which is why you are getting a lot of wtf forGG posts and rightly so.
The lack of data for Kespa pros means that people like roro innovation rain and flash aren't anywhere near as high as they should be based on how many wins they have been getting which means that they will need to continue to play dominantly for an unfairly long time to rank where they should be. This needs to be bolded on any "top ten list" which clearly doesn't show the top ten players.
What this list makes a lot of people overlook in their justified confusion is that statistically it is exciting that overall, Zerg are winning significantly less compared to 2012 and this may mean a more exciting spectator experience. So it begs the question: why post such a list?
For the purposes of a less contemptible ranking list would it be possible for it to represent a separate data set which reset either when Kespa joined sc2 or when hots came out?
On April 19 2013 00:09 Nerski wrote: Not posting any explanation of where those numbers are from, seems like a cheap advertisement for your site rather then paying TL to advertise.
I don't know man, I linked to the FAQ earlier. It contains most of what you are looking for. The source code for the website is available as well. I thought maybe I for once could just post a thread without prefacing it with copy pasta that most regular forumgoers here have read many times already. The site is ad-free and clinically stripped of ways to fatten my wallet. And that is also why I don't feel inclined to answer your second question. Feel free to use TLPD. It's a nice and accurate resource, but it doesn't offer even a fraction of the cool things one can do with the data that they have.
On April 19 2013 00:19 dreamseller wrote: For the purposes of a less contemptible ranking list would it be possible for it to represent a separate data set which reset either when Kespa joined sc2 or when hots came out?
It would be possible, but trust me on this, it would be far more WTF-worthy and confusing than ForGG on #4.
Edit: Just so it's clear, I'm fine with constructive criticism. I get it a lot and many times they are entirely right.
On April 19 2013 00:19 dreamseller wrote: I LOVE RANKING LISTS.
The sample size of this data is too small so the rankings are too far from current realities to be posting inflammatory "top ten lists" or claims that a race's dominance is over (although coincidentally it looks like that may be the case as tvz is looking reasonable at the highest level). The statistics and math is very interesting to a psyche major but also a bit disappointing that you haven't included more commmentary with this post, which is why you are getting a lot of wtf forGG posts and rightly so.
The lack of data for Kespa pros means that people like roro innovation rain and flash aren't anywhere near as high as they should be based on how many wins they have been getting which means that they will need to continue to play dominantly for an unfairly long time to rank where they should be. This needs to be bolded on any "top ten list" which clearly doesn't show the top ten players.
What this list makes a lot of people overlook in their justified confusion is that statistically it is exciting that overall, Zerg are winning significantly less compared to 2012 and this may mean a more exciting spectator experience. So it begs the question: why post such a list?
For the purposes of a less contemptible ranking list would it be possible for it to represent a separate data set which reset either when Kespa joined sc2 or when hots came out?
I think I commented on most of your issues in my post with the big-ass picture on page 3 of this thread
It's a pretty incredible effort, even if the lay of the land of SC2 will highlight some of the weaker points of such a system.
I wonder, have you ever thought about making an online/"lan" distinction? I suppose it really should have been done the first time the data was entered. Maps could also be interesting. But it would obviously be a big effort on top of an already giant one.
On April 19 2013 00:28 m0ck wrote: It's a pretty incredible effort, even if the lay of the land of SC2 will highlight some of the weaker points of such a system.
I wonder, have you ever thought about making an online/"lan" distinction? I suppose it really should have been done the first time the data was entered. Maps could also be interesting. But it would obviously be a big effort on top of an already giant one.
Best of luck, and thanks, I enjoy the site a lot.
That is already done. http://aligulac.com/results/ If there is a globe, it's online, if there are 2 persons, it's offline.
No we haven't included maps, and at this point we won't either due to the fact that we then have to plow through 120k games.
It's surprising to see so few protoss at the top. From my own experience I've been having a ton of trouble with oracle openers and their mid and late game splash damage.
Kinda hard to be attracted to the site, when the ranking system is not an actual representation of players actual rankings in accordance with their fellow pro's.
Basically according to the rankings, it is better to stomp a bunch of no-namers in a WCS qualifier, than to get top 8 in GSL... which i think pretty much every single person in the world who follows esports, will comment, one is definitely not comparable to the other.
but i do love reading threads like this where we get
"Player X is really good! Trust me I have seen him play, and it was really good!" Because clearly opinion points > He won an international tournament, beating several known koreans.... o wait no foreigner can do that >.>
also "He won X tournament" who was there? "A bunch of non-koreans" then your win doesnt hold much value when you are comparing them to someone who faces Koreans in literally every single tournament they attend.
On April 19 2013 00:36 MaestroSC wrote: Kinda hard to be attracted to the site, when the ranking system is not an actual representation of players actual rankings in accordance with their fellow pro's.
Basically according to the rankings, it is better to stomp a bunch of no-namers in a WCS qualifier, than to get top 8 in GSL... which i think pretty much every single person in the world who follows esports, will comment, one is definitely not comparable to the other.
but i do love reading threads like this where we get
"Player X is really good! Trust me I have seen him play, and it was really good!" Because clearly opinion points > He won an international tournament, beating several known koreans.... o wait no foreigner can do that >.>
also "He won X tournament" who was there? "A bunch of non-koreans" then your win doesnt hold much value when you are comparing them to someone who faces Koreans in literally every single tournament they attend.
I totally agree. The biggest issue for us, is that we have no objective way of truly ranking tournaments according to how "hard they are". Plus, since most koreans are ranked higher than foreigners. Beating them will give you more points than beating a below your level foreigner. Just saying.
Also, this site shouldn't be taken as "THE TRUTH" because there can be anomalies, like ForGG and his french-plowing madness. However, all in all, the site does do a better job at predicting results across the board than anything we know of. And the points that ForGG has earned in a closed enviroment (the same can be said for Bunny and the Danish scene) will be lost again to more skilled players who will then rise the ranks.
What I am basically saying is that ranking placement =/= pure skill. It's just a somewhat decent proxy.
On April 19 2013 00:44 crms wrote: Are protoss really that underwhelming in HOTS? Or is it residual effect from lackluster WoL performance and they need to catch up?
From the FAQ
How do the races OP/UP work?
On the period list you can see OP/UP fields, and in the infobox for each period, the same data is given as "leading" and "lagging" race. This is an indicator showing which races are most and least prominent near the top of the list. Specifically, for each race imagine a hypothetical player with a rating equal to the mean of the ratings of the top five players of that race, and imagine these three players playing very many games against each other. If the players were of equal strength, each of them would score about 50%, however, in reality, one of them may score, say, 10% more than that. The race that scores the most in this scenario is the "OP", or "leading" race, and the race that scores the least is the "UP", or "lagging" race.
This is provided as a way to analyse the metagame shifts near the top of the skill ladder, and should not be taken as actual evidence for real game imbalance.
On April 19 2013 00:49 crms wrote: That's pretty cool methodology. I was more referring to the lack of protoss players in the rankings. I like this OP/UP measure though, really neat.
Yeah, the answer to your question is yes, there is a catch-up thing going on. It's hard to say exactly how much of what you're seeing is WoL and how much is HotS, but presumably a non-negligible amount of it is still WoL.
On April 18 2013 22:52 Grovbolle wrote: ONLY WINS MATTER. Meaning that a win in an online qualifier Bo3 is the same as beating someone in Ro4 GSL.
Isn't this a big flaw of the ranking system?
That depends, can you objectively rank matches simply based on the context in which they were played?
I don't know how hard it would be to do that but i think it could fix a few problems.
Maybe a similar system to what UEFA uses for the Golden Shoe award.
Since the 1996–97 season, European Sports Media have awarded the Golden Shoe based on a points system that allows players in tougher leagues to win even if they score fewer goals than a player in a weaker league. The weightings are determined by the league's ranking on the UEFA coefficients, which in turn depend on the results of each league's clubs in European competition over the previous five seasons. Goals scored in the top five leagues according to the UEFA coefficients list are multiplied by a factor of two, and goals scored in the leagues ranked six to 21 are multiplied by 1.5.[4] Thus, goals scored in Serie A, the top Italian football league, will count for more than those scored in the weaker Welsh Premier League, its Welsh equivalent
Korea has the best leagues so if you win there you get more points. So we could have 3 tiers of competition:
1stWCS Korea,Proleague,WCS Season Finals,Blizzcon. 2nd WCS EU/NA,MLG,IEM,Dreamhack 3rd Everything else
Since the 1996–97 season, European Sports Media have awarded the Golden Shoe based on a points system that allows players in tougher leagues to win even if they score fewer goals than a player in a weaker league. The weightings are determined by the league's ranking on the UEFA coefficients, which in turn depend on the results of each league's clubs in European competition over the previous five seasons. Goals scored in the top five leagues according to the UEFA coefficients list are multiplied by a factor of two, and goals scored in the leagues ranked six to 21 are multiplied by 1.5.[4] Thus, goals scored in Serie A, the top Italian football league, will count for more than those scored in the weaker Welsh Premier League, its Welsh equivalent
Korea has the best leagues so if you win there you get more points. So we could have 3 tiers of competition:
1stWCS Korea,Proleague,WCS Season Finals,Blizzcon. 2nd WCS EU/NA,MLG,IEM,Dreamhack 3rd Everything else
Thing is though, points aren't "given" they are "stolen" from other players
Since the 1996–97 season, European Sports Media have awarded the Golden Shoe based on a points system that allows players in tougher leagues to win even if they score fewer goals than a player in a weaker league. The weightings are determined by the league's ranking on the UEFA coefficients, which in turn depend on the results of each league's clubs in European competition over the previous five seasons. Goals scored in the top five leagues according to the UEFA coefficients list are multiplied by a factor of two, and goals scored in the leagues ranked six to 21 are multiplied by 1.5.[4] Thus, goals scored in Serie A, the top Italian football league, will count for more than those scored in the weaker Welsh Premier League, its Welsh equivalent
Korea has the best leagues so if you win there you get more points. So we could have 3 tiers of competition:
1stWCS Korea,Proleague,WCS Season Finals,Blizzcon. 2nd WCS EU/NA,MLG,IEM,Dreamhack 3rd Everything else
It's not an inherently bad approach, but it has its problem within SC2: Is a WCS Korea Code A qualifier really better than WCS EU Code S? Where do you put tournaments like the TeSL (Taiwanese Esports League)? Is Blizzcon 2010 really as prestigious as WCS 2012? Or what about the Korean Weeklies vs. a random ZOTAC cup?
There are many, many ways to weigh matches in SC2, and pretty much all of them will give you a headache once you start to seriously think about them.
On April 18 2013 22:52 Grovbolle wrote: ONLY WINS MATTER. Meaning that a win in an online qualifier Bo3 is the same as beating someone in Ro4 GSL.
Isn't this a big flaw of the ranking system?
Not necessarily. That specific quote about the system is an oversimplification, because winning in GSL RO4 you are very likely to have beat an excellent player. The system would reward you for beating an excellent player, not for beating him a specific match. This is simply equivalent to holding the view that GSL is the best league because the best players play in it .
I wonder if you could extend the system with some form of backward propagation of skill, something like "TrueSkill through time". I think this could help with the KeSPA problem, where a bunch of players were added to the pool that are on a level much higher than the initially assumed 1000 points.
On April 19 2013 01:12 delHospital wrote: I wonder if you could extend the system with some form of backward propagation of skill, something like "TrueSkill through time". I think this could help with the KeSPA problem, where a bunch of players were added to the pool that are on a level much higher than the initially assumed 1000 points.
I do believe that is already used. (I think it's called forward filtering backwards smoothing or something)
Again, as stated several times already; The amount of points you get, depends on the 'level'/'score'/'rating' of the players you are playing against.
So a player winning many games against 1300 Europeans, might earn 'quite alot of point' maybe getting 1500? But if the same player won a few GSL games (vs players with 1700 points) the point climb would be steeper.
=> Winning GSL games gives you more points, because you fight better opponents. (Note: Beating Life at MLG or GSL, or even an online tournament, earns you the same amount of points; which is kind of fair?)
Question though: You get a different amount of point from beating an opponent 4-0, instead of 4-3, right?
Its good to see Terrans coming back again after an era of dominance by Infestors in WoL, but I feel bad about Protosses. Even after an expansion they still are the most underrepresented race in the proscene D:
On April 19 2013 01:15 Goolpsy wrote: Again, as stated several times already; The amount of points you get, depends on the 'level'/'score'/'rating' of the players you are playing against.
So a player winning many games against 1300 Europeans, might earn 'quite alot of point' maybe getting 1500? But if the same player won a few GSL games (vs players with 1700 points) the point climb would be steeper.
=> Winning GSL games gives you more points, because you fight better opponents. (Note: Beating Life at MLG or GSL, or even an online tournament, earns you the same amount of points; which is kind of fair?)
Question though: You get a different amount of point from beating an opponent 4-0, instead of 4-3, right?
Yes.
Actually you can gain points by losing as well.
Say that Life beats "insert random nobody" 4-3 Life was expected to beat him 4-1 (Which is called expected score here http://aligulac.com/players/3-Life/period/82/). The random nobody then gets a few points which life loses, because life underperformed compared to what was expected of him in a Bo7
Constructive criticism is fine, but the amount of people who complain about this rating for not including KeSPA players despite the reason behind this being explained (and the amount of people who essentially ask "WHAT IS THIS?" despite there being a FAQ linked at the end) is staggering.
Thanks for the hard work, TheBB and crew. As long as Life is #1, I won't complain!
Let me try to explain my view on some of these issues.
Should tournaments be weighted?
A good mathematical model has few parameters. The standard Glicko system makes do with four (period length, initial uncertainty, uncertainty floor and uncertainty decay rate). Having experimented for a while I have decided to add a fifth, the factor by which offline matches should be weighted relative to online ones. (This change is not live yet, by the way. I'm still working on it.)
If I were to start weighing tournaments it would introduce something like a hundred additional parameters. At this stage it's no longer a useful model. Some condensation is necessary. For example: you can do linear weighing by prize pool, or you can do linear weighing by the mean rating of participating players, or something like that. I'm not saying this will be useless, but conventional mathematical wisdom is to enhance your model with the parameters that are likely to yield the best results with the least expense in complexity first, and while I can't out of hand dismiss the benefits of this scheme, I can tell that it will be complex.
Also, please note that you already get more points by beating higher rated players. As such, wins in the GSL automatically count for more because you will likely face better opposition there.
Also worth keeping in mind is that if I increase the weight of a tournament, it will not only increase the point gain of the winners, but also the losses of the losers. It will not make the mean rating of Code S higher, but it would increase the gap between the best and the worst Code S players. This may not be what you actually want.
Filtering
I have already implemented backwards filtering, but we're still discussing where it makes sense to display filtered numbers, and where it makes sense to display the raw ones. Currently, only the player charts show filtered data. I was reluctant to use those in the rating lists, because then they can change after the fact, which is somewhat counterintuitive and can confuse users (and I'm not short of confused users, as you can see).
If people want to see filtered data in rating lists, then I can do that. Note that it will not change the current list though, only the past ones.
Everytime Aligulac content is posted there's a handful of really stupid comments either not reading OP or the faq, or lacking basic grasp of math. >_< Please don't let that discourage you guys, personally I find Aligulac really useful and very interesting.
On April 19 2013 01:30 Kasaraki wrote: Everytime Aligulac content is posted there's a handful of really stupid comments either not reading OP or the faq, or lacking basic grasp of math. >_< Please don't let that discourage you guys, personally I find Aligulac really useful and very interesting.
I always feel like the few of us who actively post in these threads, explaining the site etc. ends up coming off as ULTRA defensive and not willing to take criticism. (Me, Conti, TheBB and Otolia mainly).
Constructive criticism is always good, however before you start questioning the math, please remember this. TheBB is studying a PhD in math. So please bring at least a decent argument and get familiarized with Glicko rating and basic statistical knowledge as well as what the site actually is and most importantly, what it is NOT before derping out stuff like: "No KESPA = Shit rating"
The Kespa players will get their points soon enough. Flash, Innovation, Rain, and Roro are all in the top 20. Playing in the GSL/OSL/WCS will provide plenty of mixing with established ESF players to balance the ratings.
On April 19 2013 01:34 Grovbolle wrote: I always feel like the few of us who actively post in these threads, explaining the site etc. ends up coming off as ULTRA defensive and not willing to take criticism. (Me, Conti, TheBB and Otolia mainly).
Yes, please, that is not my intention at all. I love the fact that this sparks some debate.
Stats in SC are so funny because every time someone proposes some stats and interpretation someone says the sample size is too small and by the time we have a large enough sample size half of the sample is from 6 months ago and 'round and 'round we go.
That's pretty cool, I didn't know you were already doing backwards filtering.
Do you think it would change much if you seeded Koreans with higher initial ratings that foreigners? I realize that it doesn't fit nicely with the model, but would it be more accurate?
On April 19 2013 01:55 Novacute wrote: A top 10 foreigner list that does not include BabyKnight and Mana? There's definitely something wrong here.
Maybe, take your time to read the OP, the FAQ and several explanations in this thread ?
FUCK IT. IT SUCKS.
Btw, there was a time i used to "cheat" using aligulac on liquibets. It got me to top 20 easilly haha.
If a player has not played any games for four periods (eight weeks, or about two months), they will not be included.
Mana has been playing very well in the ATC, and he's been a prominent player in recent EU tournmaents like Zotac and WCS europe. Surely, that is less than eight weeks.
Edit: Found him, surprised he wasn't ranked any higher. I clearly overestimated his ability, but the way it's gauged is very arbitrary, he's been consistently winning against Koreans, shouldn't that yield a far better result than otherwise indicates?
On April 19 2013 01:55 Novacute wrote: A top 10 foreigner list that does not include BabyKnight and Mana? There's definitely something wrong here.
Maybe, take your time to read the OP, the FAQ and several explanations in this thread ?
FUCK IT. IT SUCKS.
Btw, there was a time i used to "cheat" using aligulac on liquibets. It got me to top 20 easilly haha.
If a player has not played any games for four periods (eight weeks, or about two months), they will not be included.
Mana has been playing very well in the ATC, and he's been a prominent player in recent EU tournmaents like Zotac and WCS europe. Surely, that is less than eight weeks.
Mana is on 72nd place in the world, and 17th place among foreigners. He has not been removed, he is simply not in the top 10 on account of there being limited space there (namely, 10 spots).
I hope you didn't think that only 20 people have played a game in the last two weeks. I'm not about to copy the whole list of 680 people in here.
Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
Sorry if it's a bad question, but is there a "time decay" variable factored in here? Like, older results will matter less and less than the newer ones? It could be one way to "fix" the KesPA problem (although I'm in agreement that they should, eventually, start invading the top 10). Anyway, while I'm already happy with ELO, another ranking method isn't a bad thing. It's just a bit too objective (and lacks contextual variables that to my mind are very important, but nearly impossible to mathematically compute).
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
I have access to code. I've read it and understood it. I doubt you can say the same.
The most stupid part in your post is you considering that your opinion on who should stand at the top is relevant to the accuracy of the ratings. The ratings are accurate, the formula may not be the best or the most precise but the ratings are accurate according to the system. There is nothing more to it.
So unless you actually have a worthwhile suggestion to make, you should stop being an ass to people who are dedicating a significant part of their life to a project like this one.
Edit : And so we are clear, I am not pretending that everything is right and that criticism isn't acceptable. I don't like ForGG being 4th either as it isn't reflecting his actual skill level but saying that the system overall is bad because of a single case is beyond absurd.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please remember that all our site does is track and distribute points. I refer to my post on page 3 with a big-ass picture explaining how points are distributed.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
Why not? That is a pretty big flaw in the system if it produces a ranking that has forgg in the top 5. You guys are pretty sensitive and not open to any criticism I see. A person who took their system more seriously and strove for accuracy would try to figure out why forgg got ranked so high and try to fix the flaws instead of pretending no flaws exist.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
Any post saying Forgg is not very good should not be taken seriously.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
Why not? That is a pretty big flaw in the system if it produces a ranking that has forgg in the top 5. You guys are pretty sensitive and not open to any criticism I see. A person who took their system more seriously and strove for accuracy would try to figure out why forgg got ranked so high and try to fix the flaws instead of pretending no flaws exist.
Perhaps there should be a FAQ with the most common and/or obvious "fixes" to the ranking system somewhere. Believe me when I say that a) the ranking system is not final and is being worked on, and b) many things have been considered and are being considered. And most of the obvious fixes turn out not to be so useful and/or easy once you start to think about them.
The ranking definitely does have flaws, and I'd say nobody has ever called it perfect, or even close to it. It's not. It will most likely never be, not with a game with so many random factors such as Starcraft 2. It is, however, still pretty damn accurate, compared to all the other ranking systems out there. Just try it: Go to the prediction page and use aligulac's predictions for your liquidbets. You will do significantly better than the average.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
Why not? That is a pretty big flaw in the system if it produces a ranking that has forgg in the top 5. You guys are pretty sensitive and not open to any criticism I see. A person who took their system more seriously would try to figure out why forgg got ranked so high and try to fix the flaws instead of pretending no flaws exist.
As already posted, we are all aware of exactly why forgg is ranked as highly as he is. We also explained why. We are also debating how to improve this internally, but currently haven't come up with a solution. The best current solution is to wait for him to lose points (which is inevitable). This system uses likelihood optimization, which basically means that if a player gets a better score than he was expected to, we raise the rating of the player so that the overall likelihood prediction of the entire model fits the data set available. With ForGG plowing players, 2-0'ing lots of players he was only predicted to beat, let's say 2-1, his rating improved. And because he played quite a few players, it increased a lot.
You quite simply bash something you do not understand without adding anything to the discussion, not even having the courtesy to read the thread since I clearly adressed this on page 3.
Who is that #1 Terran? Is that Lucifron, or something? That person I think finished #1 last WoL season too. That's a ridiculous amount of games with even a higher win-rate than ForGG.
Some players will always be higher than their perceived strength, this just happens with every kind of rating system.
Just like no rating system has currently the KeSPA players at their respected perceived strength (imo), there needs to be more point exchange (matches) between the different scenes (esf, kespa, international) for everything to balance in a more meaningful way.
There seems to be several posts along the lines of "I disagree with your statistics because player X has rank Y".
It would be interesting to know if those posters think it's possible to improve on the model (and if so, how) or if it's just generic internet-based bashing of things you don't agree with.
Since both the code and the data is available for download, anyone with the required skill (or funds to pay someone with the required skill) could alter the model in any way they choose. You could factor in age, shoe size and twitter followers if you wanted.
Side note; as someone who loves starcraft and is currently studying statistics and python in my spare time, this project is a god-send. Thank you and keep up the good work.
Obviously these things are not completely accurate of skill. Results don't even reflect skill very well, so it's even less likely a rating system would. Innovation is OBVIOUSLY doing better than Leenock, but he is not on the list. Don't get too caught up in these types of lists. Life's placement makes sense, though.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
Why not? That is a pretty big flaw in the system if it produces a ranking that has forgg in the top 5. You guys are pretty sensitive and not open to any criticism I see. A person who took their system more seriously and strove for accuracy would try to figure out why forgg got ranked so high and try to fix the flaws instead of pretending no flaws exist.
Perhaps there should be a FAQ with the most common and/or obvious "fixes" to the ranking system somewhere. Believe me when I say that a) the ranking system is not final and is being worked on, and b) many things have been considered and are being considered. And most of the obvious fixes turn out not to be so useful and/or easy once you start to think about them.
The ranking definitely does have flaws, and I'd say nobody has ever called it perfect, or even close to it. It's not. It will most likely never be, not with a game with so many random factors such as Starcraft 2. It is, however, still pretty damn accurate, compared to all the other ranking systems out there. Just try it: Go to the prediction page and use aligulac's predictions for your liquidbets. You will do significantly better than the average.
Well at least you can take criticism and admit the system has flaws, so thanks for that. I will agree that sc2 is too volatile to produce a very accurate list, especially with bo3s mainly, since I think a bo5 or 7 would be more accurate but that's not practical for most tournaments. I'm not saying your system is awful, but was just pointing out one of the more serious flaws in the system.
On April 19 2013 03:11 Blargh wrote: Obviously these things are not completely accurate of skill. Results don't even reflect skill very well, so it's even less likely a rating system would. Innovation is OBVIOUSLY doing better than Leenock, but he is not on the list. Don't get too caught up in these types of lists. Life's placement makes sense, though.
He is getting there quickly though. Currently 14th in world, 13th Korea.
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
Why not? That is a pretty big flaw in the system if it produces a ranking that has forgg in the top 5. You guys are pretty sensitive and not open to any criticism I see. A person who took their system more seriously and strove for accuracy would try to figure out why forgg got ranked so high and try to fix the flaws instead of pretending no flaws exist.
Perhaps there should be a FAQ with the most common and/or obvious "fixes" to the ranking system somewhere. Believe me when I say that a) the ranking system is not final and is being worked on, and b) many things have been considered and are being considered. And most of the obvious fixes turn out not to be so useful and/or easy once you start to think about them.
The ranking definitely does have flaws, and I'd say nobody has ever called it perfect, or even close to it. It's not. It will most likely never be, not with a game with so many random factors such as Starcraft 2. It is, however, still pretty damn accurate, compared to all the other ranking systems out there. Just try it: Go to the prediction page and use aligulac's predictions for your liquidbets. You will do significantly better than the average.
Well at least you can take criticism and admit the system has flaws, so thanks for that. I will agree that sc2 is too volatile to produce a very accurate list, especially with bo3s mainly, since I think a bo5 or 7 would be more accurate but that's not practical for most tournaments. I'm not saying your system is awful, but was just pointing out one of the more serious flaws in the system.
Well, there is a difference between pointing out flaws in the system (there are) and saying "Your system sucks and should be ignored".
On April 19 2013 02:20 Canucklehead wrote: Errr I dunno what polt has done lately to be ranked number 2. And forgg 4th? He's not very good. I dunno if he's still 1-1-1 against every race in Hots, like he did in Wol though. Any list that has forgg in the top 5 shouldn't be taken seriously. These lists are an interesting talking point, but not that accurate I find. Except for life. He's the number 1 bonjwa in the scene right now.
This is not a list, it's a ranking. It's not an appraisal or an opinion but the result of a scientifically correct representation of the players skill. Unless you have mathematically correct definition of accuracy in this present case, what you said is just an opinion and a not very forgiving one at that.
You must not be a very critical thinker and take things at face value. If you were you would know the ranking is based on a formula made by a human and is not infallible. I'm saying the formula used to come up with this ranking has flaws in it. I'm not sure why you think it produces an accurate ranking just because it uses fancy stats.
Do you take any list a statistician comes up with as fact without questioning the methods behind them? That's a serious question, since you just assumed everything was a "scientifically correct representation of the players skill" because he told you it was.
Please don't just come in bashing the entire project because ForGG is pretty highly rated.
Why not? That is a pretty big flaw in the system if it produces a ranking that has forgg in the top 5. You guys are pretty sensitive and not open to any criticism I see. A person who took their system more seriously and strove for accuracy would try to figure out why forgg got ranked so high and try to fix the flaws instead of pretending no flaws exist.
Perhaps there should be a FAQ with the most common and/or obvious "fixes" to the ranking system somewhere. Believe me when I say that a) the ranking system is not final and is being worked on, and b) many things have been considered and are being considered. And most of the obvious fixes turn out not to be so useful and/or easy once you start to think about them.
The ranking definitely does have flaws, and I'd say nobody has ever called it perfect, or even close to it. It's not. It will most likely never be, not with a game with so many random factors such as Starcraft 2. It is, however, still pretty damn accurate, compared to all the other ranking systems out there. Just try it: Go to the prediction page and use aligulac's predictions for your liquidbets. You will do significantly better than the average.
Well at least you can take criticism and admit the system has flaws, so thanks for that. I will agree that sc2 is too volatile to produce a very accurate list, especially with bo3s mainly, since I think a bo5 or 7 would be more accurate but that's not practical for most tournaments. I'm not saying your system is awful, but was just pointing out one of the more serious flaws in the system.
A players ranking is not a flaw in the system. A flaw in the system would be talking about a specific flaw in their methodology.
These threads always descend into people criticising it based on who is actually ranked highly. This system will get more accurate with time. HoTs just came out. Give it time. It will be refined and sample sizes will increase.
Some of the reception in this thread is absolutely disgusting. I actually think some people just hate everything to do with math and statistics and refuse, no matter what, to accept it, or appreciate it, or understand it. They just reject it wholly at face value because it's math and because I DISAGREE WITH IT (even despite explanation).
If you're super hung up on the fact that the system has flaws, you shouldn't be, because every mathematical model in existence has flaws. It's nothing new, and nothing that the guys at Aligulac weren't already well aware of before the site was even online months ago.
If you're super hung up on the fact that some of the ratings are weird, maybe you should go look at the TLPD ELO list and see if you can rationalize mOOnGLade and DeMusliM being 3rd and 4th out of all foreigners, and Curious being better than RorO with Flash/Innovation both not in the top 5 at all? Making a rating system is really god damn hard, and the Aligulac one already blows the TLPD one out of the water in many regards. Expecting someone to make a perfect rating system is very unreasonable, especially because anyone who's seen an LR thread in their life can tell you that people have different expectations of 'perfect' anyway - some want a system that calls the best player the one with the most long term accomplishments, while some want a system that calls the best player the one who killed the most nerds in the last 48 hours. No matter what someone will call it imperfect.
And if you feel like ratings are pointless for a game like SC2... okay? If you feel the statistics hold no value or intrigue, then don't read them. It's not like anyone is forcing you to, and crusading the point as if nobody cares is just wrong, because there is people in the LR threads every week who post and talk about the Aligulac predictions. Statistics like this exist for interest and consideration, not to be followed like a church. If you aren't interested or feel they have no value, then you're free to ignore them.
Also, TheBB used the system to do his liquibetting and is (was?) leading in winrate. So technically it's already done better than most people's subjective analysis of player skill, despite its flaws.
On April 19 2013 05:02 MCXD wrote: Also the reception in this thread is absolutely disgusting. I actually think people just hate everything to do with math and statistics and refuse, no matter what, to accept it, or appreciate it, or understand it. They just reject it wholly at face value, because it's math and because I DISAGREE WITH IT.
Anyway, TheBB used the system to do his liquibetting last season, and won (highest win rate). So technically it's done better than most people's supposed subjective analysis of player skill.
I have to say that this fact carrys HUGE weight. If a mathetical model is extremely accurate in its prediction power, that says a LOT about the model imo.
Keep up the good work. All this is very interesting indeed.
On April 19 2013 05:02 MCXD wrote: Also the reception in this thread is absolutely disgusting. I actually think people just hate everything to do with math and statistics and refuse, no matter what, to accept it, or appreciate it, or understand it. They just reject it wholly at face value, because it's math and because I DISAGREE WITH IT.
Anyway, TheBB used the system to do his liquibetting last season, and won (highest win rate). So technically it's done better than most people's supposed subjective analysis of player skill.
I have to say that this fact carrys HUGE weight. If a mathetical model is extremely accurate in its prediction power, that says a LOT about the model imo.
Keep up the good work. All this is very interesting indeed.
Except it kept saying Welmu would qualify for WCS EU! (okay.. "should") + Show Spoiler +
On April 19 2013 05:02 MCXD wrote: Anyway, TheBB used the system to do his liquibetting last season, and won. So technically it's already done better than most people's subjective analysis of player skill.
Don't Liquibet winners get some special icon next to their name (TheBB doesn't have one)?
On April 19 2013 05:02 MCXD wrote: Anyway, TheBB used the system to do his liquibetting last season, and won. So technically it's already done better than most people's subjective analysis of player skill.
Don't Liquibet winners get some special icon next to their name (TheBB doesn't have one)?
I don't actually understand the season structure of liquibet so don't ask me, haha. I'll edit my post to be less bad. Turns out I was slightly remembering wrong anyway.
I think it might actually be this season that's still going on.
On April 19 2013 05:02 MCXD wrote: Anyway, TheBB used the system to do his liquibetting last season, and won. So technically it's already done better than most people's subjective analysis of player skill.
Don't Liquibet winners get some special icon next to their name (TheBB doesn't have one)?
Not sure i thought special icons came from having lots of posts and/or interesting posts, not sure though :D
I really wish you would of seperated HOTS and WoL results...so much of this is still coming off of WoL, so quite frankly I think it means nothing until we get more results in HOTS and WoL is diluted out.
On April 19 2013 05:02 MCXD wrote: Anyway, TheBB used the system to do his liquibetting last season, and won. So technically it's already done better than most people's subjective analysis of player skill.
Don't Liquibet winners get some special icon next to their name (TheBB doesn't have one)?
I don't actually understand the season structure of liquibet so don't ask me, haha. I'll edit my post to be less bad.
I think it might actually be this season that's still going on.
On April 19 2013 05:33 Prplppleatr wrote: I really wish you would of seperated HOTS and WoL results...so much of this is still coming off of WoL, so quite frankly I think it means nothing until we get more results in HOTS and WoL is diluted out.
If we did, then the ratings would have NO validity what so ever. We have less than 2K hots matches.
Someone already linked TLPD HotS, which has Empire Happy as the best player IN THE WORLD
On April 19 2013 05:33 Prplppleatr wrote: I really wish you would of seperated HOTS and WoL results...so much of this is still coming off of WoL, so quite frankly I think it means nothing until we get more results in HOTS and WoL is diluted out.
If we did, then the ratings would have NO validity what so ever. We have less than 2K hots matches.
Someone already linked TLPD HotS, which has Empire Happy as the best player IN THE WORLD
That's my point. We don't have enough results...so until we can get more results and dilute WoL out, it really has no meaning as to the current state.
On April 19 2013 05:33 Prplppleatr wrote: I really wish you would of seperated HOTS and WoL results...so much of this is still coming off of WoL, so quite frankly I think it means nothing until we get more results in HOTS and WoL is diluted out.
If we did, then the ratings would have NO validity what so ever. We have less than 2K hots matches.
Someone already linked TLPD HotS, which has Empire Happy as the best player IN THE WORLD
That's my point. We don't have enough results...so until we can get more results and dilute WoL out, it really has no meaning as to the current state.
So you just want us to make no prediction whatsoever in 6 months?
If so, then just ignore anything we do for the next couple of months, then when we have a lot of results, we can discuss if we want to separate the two
On April 19 2013 05:33 Prplppleatr wrote: I really wish you would of seperated HOTS and WoL results...so much of this is still coming off of WoL, so quite frankly I think it means nothing until we get more results in HOTS and WoL is diluted out.
If we did, then the ratings would have NO validity what so ever. We have less than 2K hots matches.
Someone already linked TLPD HotS, which has Empire Happy as the best player IN THE WORLD
That's my point. We don't have enough results...so until we can get more results and dilute WoL out, it really has no meaning as to the current state.
So you just want us to make no prediction whatsoever in 6 months?
If so, then just ignore anything we do for the next couple of months, then when we have a lot of results, we can discuss if we want to separate the two
LOL, what? I never said anything to that point. What you guys are doing is great. I'm simply pointing out the obvious, to those who can't recognize it, so that they don't take this as such fact. It is statistics based on a lack of information for the state of HOTS. Make all the predictions you want (I already have for WCS NA), but it should be warned (as I am doing) that these are less reliable than they were in WoL (because of the lack of information).
IE. statistics are great, but they are dangerous - especially when based on a lack of information. People should take this with more of a grain of salt.
@conti, agreed. However, the reason I said that I wanted to see a separation is because it would make people confused and more inclined to see why (look at details). Then realize that there is a lack of information, which is the truth. I have nothing against you guys for making a post to say the leading race (based on your site) has changed, but people need to realize the details of the situation and not make such crazy assumptions and interpretations (already one closed post because of it, lol). Again, great job to everyone at aligulac, appreciate it, I didn't mean any offense or anything negative to what you are doing. I just know that some people never bother with details, so I try to add information so they may realize it.
On April 19 2013 05:02 MCXD wrote: Anyway, TheBB used the system to do his liquibetting last season, and won. So technically it's already done better than most people's subjective analysis of player skill.
Don't Liquibet winners get some special icon next to their name (TheBB doesn't have one)?
Not sure i thought special icons came from having lots of posts and/or interesting posts, not sure though :D
On April 19 2013 05:33 Prplppleatr wrote: I really wish you would of seperated HOTS and WoL results...so much of this is still coming off of WoL, so quite frankly I think it means nothing until we get more results in HOTS and WoL is diluted out.
If we did, then the ratings would have NO validity what so ever. We have less than 2K hots matches.
Someone already linked TLPD HotS, which has Empire Happy as the best player IN THE WORLD
That's my point. We don't have enough results...so until we can get more results and dilute WoL out, it really has no meaning as to the current state.
So you just want us to make no prediction whatsoever in 6 months?
If so, then just ignore anything we do for the next couple of months, then when we have a lot of results, we can discuss if we want to separate the two
LOL, what? I never said anything to that point. What you guys are doing is great. I'm simply pointing out the obvious, to those who can't recognize it, so that they don't take this as such fact. It is statistics based on a lack of information for the state of HOTS. Make all the predictions you want (I already have for WCS NA), but it should be warned (as I am doing) that these are less reliable than they were in WoL (because of the lack of information).
IE. statistics are great, but they are dangerous - especially when based on a lack of information. People should take this with more of a grain of salt.
People should. They should take any kind of statistics with more of a grain of salt, in fact.
But as for this ranking and HotS, keeping the WoL games in proves to offer a far more reliable ranking than taking just the HotS games themselves. There will always be a new expansion, or a new patch, or a new metagame, that will make most of the previous games "obsolete", so to speak. But, for the most part, these players are good (or bad) because of their skill, and not because of WoL or the current patch version.
On April 19 2013 05:02 MCXD wrote: Also, TheBB used the system to do his liquibetting and is (was?) leading in winrate. So technically it's already done better than most people's subjective analysis of player skill, despite its flaws.
For a while I had a winrate roughly equal to the leader of the LB rank. I was never actually leading since I jumped into the season late. That blog was posted before the Up and Downs, where we didn't do so well. Another guy posted earlier in this thread that he used Aligulac to get to top 20 with no problems, which agrees with what I found. I didn't keep it up (too lazy/irregular) so I will not win LB, now or ever.
So in its current state, I feel very few can legitimately claim that Aligulac tells them nothing they didn't already know. But for sure it's possible to beat it if you know what you're doing.
What if we could get 50 of us fans to pledge $10 each for a best of 7 show match between ForGG and Flash? Someone awesome like Husky would probably cast it for free. That could help move points from EU to Kespa and would be fun to watch.
Hmm, Bisu is currently at a measly 213st place, due to him not doing all too well in WoL. He seems to be pretty good in HotS though, and he sure as hell is better than a lot of those 212 other dudes.
On April 19 2013 06:29 Salient wrote: What if we could get 50 of us fans to pledge $10 each for a best of 7 show match between ForGG and Flash? Someone awesome like Husky would probably cast it for free. That could help move points from EU to Kespa and would be fun to watch.
Not sure if it is that easy, to get a kespa player to something like that. Especially with Proleague and Gsl ongoing, i don't see that happening.
On April 19 2013 06:47 Conti wrote: Hmm, Bisu is currently at a measly 213st place, due to him not doing all too well in WoL. He seems to be pretty good in HotS though, and he sure as hell is better than a lot of those 212 other dudes.
On April 18 2013 22:52 Grovbolle wrote: ONLY WINS MATTER. Meaning that a win in an online qualifier Bo3 is the same as beating someone in Ro4 GSL.
Isn't this a big flaw of the ranking system?
That depends, can you objectively rank matches simply based on the context in which they were played?
That would be similar to how teh old MTG system worked, different lvel of tournament had a different "K" value, meaning in bigger tournament the points swings would be bigger, while a pre release event would be K8 with a max of 8 points swing and only points for evenly matched players, a noral event would be k16, a GP k32, worlds k40.
On April 19 2013 06:47 Conti wrote: Hmm, Bisu is currently at a measly 213st place, due to him not doing all too well in WoL. He seems to be pretty good in HotS though, and he sure as hell is better than a lot of those 212 other dudes.
Also, it's Bisu. So I vote Bisu!
If he really does well, he will clim dont sorry, rankings are not an instant measurement of skill, you have to earn it
Well that list looks absoutely inaccurate considering the koreans that are on it. Life is not surprising but the rest are. I wouldnt use this data for anything, TLPD ELO is way better.
I don't think you should be counting online games, and also beating players with a much lower rating than you shouldn't count as much, I think ForGG at #4 is proof of these 2 issues
On April 19 2013 08:45 IcedBacon wrote: Well that list looks absoutely inaccurate considering the koreans that are on it. Life is not surprising but the rest are. I wouldnt use this data for anything, TLPD ELO is way better.
It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that ranked forGG as the 4th best player in the world I would be too embarrassed to post these statistics anywhere. Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
On April 19 2013 17:04 ninazerg wrote: According to your statistics, there are 5 zergs in the Korean top 10, 1 protoss, and 4 terrans. How is that "Zerg rule finally toppled"?
The mean rating of the top five Terrans is higher than the mean rating of the top five Zergs. It's in the FAQ.
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
*standing ovations*
Yeah that is not what I was saying at all. I simply stated that while our system is far from perfect, it does have a higher predictive power than eg. Liquibet players, and that is without taking stuff like map balance in to consideration.
I am glad that you are able to rank players with your own perfect knowledge of their skill level. We simply try to do based on whatever results are available to us from various tournaments while also building a system which is also capable of predicting outcomes of games.
As always, if you are capable of making a better system, we encourage you to do so, we are already doing all the hard work, which is collecting and organizing the matches as well as categorizing stuff like online/offline and WoL/HotS and making it available to the public so they themselves can experiment with it. If not, then please give suggestions as to what can be improved regarding the ranking method and how you propose we do that.
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
*standing ovations*
Yeah that is not what I was saying at all. I simply stated that while our system is far from perfect, it does have a higher predictive power than eg. Liquibet players, and that is without taking stuff like map balance in to consideration.
I am glad that you are able to rank players with your own perfect knowledge of their skill level. We simply try to do based on whatever results are available to us from various tournaments while also building a system which is also capable of predicting outcomes of games.
As always, if you are capable of making a better system, we encourage you to do so, we are already doing all the hard work, which is collecting and organizing the matches as well as categorizing stuff like online/offline and WoL/HotS and making it available to the public so they themselves can experiment with it. If not, then please give suggestions as to what can be improved regarding the ranking method and how you propose we do that.
The fact that nobody has come up with a better system doesn't justify the fact that your system is inaccurate.
You seem like the kind of person that would say things like.. "Yes I failed the test, but so did my best friend and he had even less points so it's all good!" Except for the fact that you failed of course.
There is no point in doing statistics poorly. Do your statistics have any descriptive power? Hardly. Do they have any prescriptive power? Absolutely not.
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
*standing ovations*
Yeah that is not what I was saying at all. I simply stated that while our system is far from perfect, it does have a higher predictive power than eg. Liquibet players, and that is without taking stuff like map balance in to consideration.
I am glad that you are able to rank players with your own perfect knowledge of their skill level. We simply try to do based on whatever results are available to us from various tournaments while also building a system which is also capable of predicting outcomes of games.
As always, if you are capable of making a better system, we encourage you to do so, we are already doing all the hard work, which is collecting and organizing the matches as well as categorizing stuff like online/offline and WoL/HotS and making it available to the public so they themselves can experiment with it. If not, then please give suggestions as to what can be improved regarding the ranking method and how you propose we do that.
The fact that nobody has come up with a better system doesn't justify the fact that your system is inaccurate.
You seem like the kind of person that would say things like.. "Yes I failed the test, but so did my best friend and he had even less points so it's all good!" Except for the fact that you failed of course.
Please point out why our system is inaccurate, other than the fact that ForGG is higher ranked than you think he should be.
From the FAQ of the site
How predictive is the system????
By my understanding, very.
This is a plot of more than 100k historical games (the whole database as of February 2013). On the horizontal axis you find predicted winrate for the presumed stronger player, using the ratings at the time the game was played. The games were grouped in reasonably small groups, i.e. 50%-53.3% and so on upwards. (Obviously no numbers below 50% since this is the predicted winrate for the stronger player.)
On the vertical axis is the actual winrate for each group.
As can be plainly seen, the actual winrate is close to the predicted winrate up to about 80%. The weighted linear fit (blue dashed line) is almost identical to the ideal curve (red dashed line) corresponding to the relationship that predicted winrate equals actual winrate.
Above 80% the winrates become a bit more erratic, and it seems like the system slightly overestimates the chances for the stronger player.
Thus, if the prediction system shows a winrate of
80% or higher in a best of 1 90% or higher in a best of 3 94% or higher in a best of 5 96% or higher in a best of 7 the numbers should be taken with a grain of salt.
You seem to just want to shit all over this without offering any valuable input, without taking the time to understand what the issues are in general just seem to be mad because whatever favorite player you might have isn't on top. Please post your Top 50 of players in the world with exact reasons as to why everyone is placed where they are.
Any ranking of playres through pure numbers, given the current environment of pro-gaming, is pointless at best and misleading at worst. You just can't do it. There's not enough sample size and not enough cross-matches among different regions to make such a ranking worthwhile.
I don't care how logical or how strong your formula is. It still doesn't mean jack squat if you lack the data to make proper conclusions. It would be like me taking the statistics of the first 5 NBA games of the season and try to come to some conclusion. When the only conclusion one can say at the time is that there can be no solid conclusion.
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
*standing ovations*
Yeah that is not what I was saying at all. I simply stated that while our system is far from perfect, it does have a higher predictive power than eg. Liquibet players, and that is without taking stuff like map balance in to consideration.
I am glad that you are able to rank players with your own perfect knowledge of their skill level. We simply try to do based on whatever results are available to us from various tournaments while also building a system which is also capable of predicting outcomes of games.
As always, if you are capable of making a better system, we encourage you to do so, we are already doing all the hard work, which is collecting and organizing the matches as well as categorizing stuff like online/offline and WoL/HotS and making it available to the public so they themselves can experiment with it. If not, then please give suggestions as to what can be improved regarding the ranking method and how you propose we do that.
The fact that nobody has come up with a better system doesn't justify the fact that your system is inaccurate.
You seem like the kind of person that would say things like.. "Yes I failed the test, but so did my best friend and he had even less points so it's all good!" Except for the fact that you failed of course.
Please point out why our system is inaccurate, other than the fact that ForGG is higher ranked than you think he should be.
This is a plot of more than 100k historical games (the whole database as of February 2013). On the horizontal axis you find predicted winrate for the presumed stronger player, using the ratings at the time the game was played. The games were grouped in reasonably small groups, i.e. 50%-53.3% and so on upwards. (Obviously no numbers below 50% since this is the predicted winrate for the stronger player.)
On the vertical axis is the actual winrate for each group.
As can be plainly seen, the actual winrate is close to the predicted winrate up to about 80%. The weighted linear fit (blue dashed line) is almost identical to the ideal curve (red dashed line) corresponding to the relationship that predicted winrate equals actual winrate.
Above 80% the winrates become a bit more erratic, and it seems like the system slightly overestimates the chances for the stronger player.
Thus, if the prediction system shows a winrate of
80% or higher in a best of 1 90% or higher in a best of 3 94% or higher in a best of 5 96% or higher in a best of 7 the numbers should be taken with a grain of salt.
You seem to just want to shit all over this without offering any valuable input, without taking the time to understand what the issues are in general just seem to be mad because whatever favorite player you might have isn't on top. Please post your Top 50 of players in the world with exact reasons as to why everyone is placed where they are.
I don't have to look into the formula you are using to be able to tell you that it doesn't work. If your formula ranks forGG as the 4th best player in the world it is simply not an accurate formula. How you got this result is irrelevant (be it due to a lack of data, a weak formula, etc. it doesn't matter), what ultimately matters is the fact that it is very inaccurate.
I don't have to know the molecular structure of fecal matter to be able to tell you that it's shite. Which is basically the argument you are trying to make.
"You don't understand the molecular structure of fecal matter so how can you be sure it's shite?"
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
*standing ovations*
Yeah that is not what I was saying at all. I simply stated that while our system is far from perfect, it does have a higher predictive power than eg. Liquibet players, and that is without taking stuff like map balance in to consideration.
I am glad that you are able to rank players with your own perfect knowledge of their skill level. We simply try to do based on whatever results are available to us from various tournaments while also building a system which is also capable of predicting outcomes of games.
As always, if you are capable of making a better system, we encourage you to do so, we are already doing all the hard work, which is collecting and organizing the matches as well as categorizing stuff like online/offline and WoL/HotS and making it available to the public so they themselves can experiment with it. If not, then please give suggestions as to what can be improved regarding the ranking method and how you propose we do that.
The fact that nobody has come up with a better system doesn't justify the fact that your system is inaccurate.
You seem like the kind of person that would say things like.. "Yes I failed the test, but so did my best friend and he had even less points so it's all good!" Except for the fact that you failed of course.
Please point out why our system is inaccurate, other than the fact that ForGG is higher ranked than you think he should be.
From the FAQ of the site
How predictive is the system????
By my understanding, very.
This is a plot of more than 100k historical games (the whole database as of February 2013). On the horizontal axis you find predicted winrate for the presumed stronger player, using the ratings at the time the game was played. The games were grouped in reasonably small groups, i.e. 50%-53.3% and so on upwards. (Obviously no numbers below 50% since this is the predicted winrate for the stronger player.)
On the vertical axis is the actual winrate for each group.
As can be plainly seen, the actual winrate is close to the predicted winrate up to about 80%. The weighted linear fit (blue dashed line) is almost identical to the ideal curve (red dashed line) corresponding to the relationship that predicted winrate equals actual winrate.
Above 80% the winrates become a bit more erratic, and it seems like the system slightly overestimates the chances for the stronger player.
Thus, if the prediction system shows a winrate of
80% or higher in a best of 1 90% or higher in a best of 3 94% or higher in a best of 5 96% or higher in a best of 7 the numbers should be taken with a grain of salt.
You seem to just want to shit all over this without offering any valuable input, without taking the time to understand what the issues are in general just seem to be mad because whatever favorite player you might have isn't on top. Please post your Top 50 of players in the world with exact reasons as to why everyone is placed where they are.
I don't have to look into the formula you are using to be able to tell you that it doesn't work. If your formula ranks forGG as the 4th best player in the world it is simply not an accurate formula. How you got this result is irrelevant (be it due to a lack of data, a weak formula, etc. it doesn't matter), what ultimately matters is the fact that it is very inaccurate.
I don't have to know the molecular structure of fecal matter to be able to tell you that it's shite. Which is basically the argument you are trying to make.
"You don't understand the molecular structure of fecal matter so how can you be sure it's shite?"
I assume that in the bolded part you mean system, and not formula
I am not really going to discuss this with you further. I agree that there are several major issues with doing statistics based on SC2 especially due to the volatility and lack of cross-region play in tournaments. Our system is simply an attempt, and if you think it's shit, I can't really change your mind.
I hope that you in general isn't as big of an asshole as you appear to be in this thread though.
On April 19 2013 18:10 baubo wrote: Any ranking of playres through pure numbers, given the current environment of pro-gaming, is pointless at best and misleading at worst. You just can't do it. There's not enough sample size and not enough cross-matches among different regions to make such a ranking worthwhile.
I don't care how logical or how strong your formula is. It still doesn't mean jack squat if you lack the data to make proper conclusions. It would be like me taking the statistics of the first 5 NBA games of the season and try to come to some conclusion. When the only conclusion one can say at the time is that there can be no solid conclusion.
It seems more like taking Euroleague stats and putting them on the same level as the NBA. Whatever results put forgg and co. so high are weighted far too strongly.
SlixSC:The reason ForGG is on #4 is not because "the formula is inaccurate" (although it could be). It's because the data is segregated. I could tweak the system so that ForGG is not on fourth place, but that might not satisfy you because your criticism is so extremely specific that it's totally useless.
The problem is that you're not being constructive. You're only pointing out problems without offering solutions, and I'm tired of it.
On April 19 2013 18:10 baubo wrote: Any ranking of playres through pure numbers, given the current environment of pro-gaming, is pointless at best and misleading at worst. You just can't do it. There's not enough sample size and not enough cross-matches among different regions to make such a ranking worthwhile.
I don't care how logical or how strong your formula is. It still doesn't mean jack squat if you lack the data to make proper conclusions. It would be like me taking the statistics of the first 5 NBA games of the season and try to come to some conclusion. When the only conclusion one can say at the time is that there can be no solid conclusion.
It seems more like taking Euroleague stats and putting them on the same level as the NBA. Whatever results put forgg and co. so high are weighted far too strongly.
Your point carries no value because you haven't even taken the time to read the numerous responses explaining why that is.
On April 19 2013 18:57 TheBB wrote: SlixSC:The reason ForGG is on #4 is not because "the formula is inaccurate" (although it could be). It's because the data is segregated. I could tweak the system so that ForGG is not on fourth place, but that might not satisfy you because your criticism is so extremely specific that it's totally useless.
The problem is that you're not being constructive. You're only pointing out problems without offering solutions, and I'm tired of it.
Because all offered solutions by other people were routinely ignored.
On April 19 2013 18:57 TheBB wrote: SlixSC:The reason ForGG is on #4 is not because "the formula is inaccurate" (although it could be). It's because the data is segregated. I could tweak the system so that ForGG is not on fourth place, but that might not satisfy you because your criticism is so extremely specific that it's totally useless.
The problem is that you're not being constructive. You're only pointing out problems without offering solutions, and I'm tired of it.
Because all offered solutions by other people were routinely ignored.
And thanks for calling me an asshole.
I'll show myself out.
No one has actually offered any solutions, just unbased criticism. The link below is Eivind's response to most of the ideas posted here, and his explanation why he thinks it is a bad idea.
We are actively testing stuff being proposed, but overall most suggestions makes the ranking both more complex, and worse. When I say worse, I mean that the predictive power of the overall system falls. This is also why we make everything available for people to DL so they can tamper with it themselves if they have an idea they want to test out.
Again, we are very open to criticism, but if you offer criticism, be prepared to also receive our justification as to why most suggestions doesn't add anything to the ranking other than making it worse.
Also: Sorry about the A-hole comment, I'll try to restrain myself from doing that.
I guess really some of these oddities just need some more interaction to sort themselves out. I guess doing statistics on these very distinctly seperated groups of people with poor mixing is a bit of a challenge.
Hmm damn, system would certainly beat my 57% liquibet prediction rate.
Is it possible that if populations A and B were mostly segregated (that is, did not play many games against each other), but pop A was significantly better than pop B, that the formula would fail to reflect this adequately? It seems to me that it would rank the top players from each population similarly, despite a skill difference. Perhaps there is enough cross-pollination between all players in real life to prevent this sort of problem?
On April 20 2013 01:46 MannerMan wrote: Is it possible that if populations A and B were mostly segregated (that is, did not play many games against each other), but pop A was significantly better than pop B, that the formula would fail to reflect this adequately? It seems to me that it would rank the top players from each population similarly, despite a skill difference. Perhaps there is enough cross-pollination between all players in real life to prevent this sort of problem?
In theory yes, this is however not a huge issue although it does create anomalies, but as time passes it usually stops being an issue. We are seeing it with kespa now, but as more of them enter gsl it evens out. I made a post on page 3 on this thread explaining it with a nice drawing
Its an awesome system with excellent predictive power. These whiney douchebags have nothing to offer except for "derp ForGG too high". You have to look at the whole system and its overall predictive value to judge how accurate a model it is, and judging by that graph and ppl using it in LB to great success, it is a very good system. Keep up the good work TheBB and others, very impressive.
On April 19 2013 18:57 TheBB wrote: SlixSC:The reason ForGG is on #4 is not because "the formula is inaccurate" (although it could be). It's because the data is segregated. I could tweak the system so that ForGG is not on fourth place, but that might not satisfy you because your criticism is so extremely specific that it's totally useless.
The problem is that you're not being constructive. You're only pointing out problems without offering solutions, and I'm tired of it.
Not sure if it is done already. But have you considered adding different factors of point swings depending on the enviroment the matches are played (tiers of tournaments) like I stated about MTG before?
On April 19 2013 17:01 SlixSC wrote: It doesn't matter that you know your system is inaccurate and somehow use that as a self-justifying argument for posting these statistics anyway. If I were to come up with a system that would show forGG as the 4th best players in the world I would be so embarrassed and definitely NOT post them anywhere on a public forum.
Statistics done poorly by people who are aware of the flaws in their system but post them anyway and make it sound like they have any descriptive power.
I don't even...
Actually, in general, over a large period of time, the aligulac system have better predictive power than the average liquibet players as well as better than any system currently out there.
Our ranking is not the ultimate truth, nor has it ever been. We are always working on making it better, and one of the issues we are very aware of are the problem with "closed pools" like the ones forGG have getting his surge in points from.
However just because it has flaws, it doesn't mean that you can't still use it, and with some common sense and supplementing with your own knowledge of the scene, try to make a fair who is the best in the world opinion.
Actually aligulac should probably be seen as a site which rewards players who have been playing good in a longer/shorter period of time and are therefore rewarded with points from players who have underperformed.
As always, I encourage people with statistical knowledge to download our DB mysql dump and try out building their own models so we can get a healthy discussion and ultimately, hopefully, improve the entire system.
Being more statistically accurate than TLPD is like taking candy from a baby.
You have found a terrible system and made a slightly better one.
*standing ovations*
Yeah that is not what I was saying at all. I simply stated that while our system is far from perfect, it does have a higher predictive power than eg. Liquibet players, and that is without taking stuff like map balance in to consideration.
I am glad that you are able to rank players with your own perfect knowledge of their skill level. We simply try to do based on whatever results are available to us from various tournaments while also building a system which is also capable of predicting outcomes of games.
As always, if you are capable of making a better system, we encourage you to do so, we are already doing all the hard work, which is collecting and organizing the matches as well as categorizing stuff like online/offline and WoL/HotS and making it available to the public so they themselves can experiment with it. If not, then please give suggestions as to what can be improved regarding the ranking method and how you propose we do that.
The fact that nobody has come up with a better system doesn't justify the fact that your system is inaccurate.
You seem like the kind of person that would say things like.. "Yes I failed the test, but so did my best friend and he had even less points so it's all good!" Except for the fact that you failed of course.
Please point out why our system is inaccurate, other than the fact that ForGG is higher ranked than you think he should be.
From the FAQ of the site
How predictive is the system????
By my understanding, very.
This is a plot of more than 100k historical games (the whole database as of February 2013). On the horizontal axis you find predicted winrate for the presumed stronger player, using the ratings at the time the game was played. The games were grouped in reasonably small groups, i.e. 50%-53.3% and so on upwards. (Obviously no numbers below 50% since this is the predicted winrate for the stronger player.)
On the vertical axis is the actual winrate for each group.
As can be plainly seen, the actual winrate is close to the predicted winrate up to about 80%. The weighted linear fit (blue dashed line) is almost identical to the ideal curve (red dashed line) corresponding to the relationship that predicted winrate equals actual winrate.
Above 80% the winrates become a bit more erratic, and it seems like the system slightly overestimates the chances for the stronger player.
Thus, if the prediction system shows a winrate of
80% or higher in a best of 1 90% or higher in a best of 3 94% or higher in a best of 5 96% or higher in a best of 7 the numbers should be taken with a grain of salt.
You seem to just want to shit all over this without offering any valuable input, without taking the time to understand what the issues are in general just seem to be mad because whatever favorite player you might have isn't on top. Please post your Top 50 of players in the world with exact reasons as to why everyone is placed where they are.
I don't have to look into the formula you are using to be able to tell you that it doesn't work. If your formula ranks forGG as the 4th best player in the world it is simply not an accurate formula. How you got this result is irrelevant (be it due to a lack of data, a weak formula, etc. it doesn't matter), what ultimately matters is the fact that it is very inaccurate.
I don't have to know the molecular structure of fecal matter to be able to tell you that it's shite. Which is basically the argument you are trying to make.
"You don't understand the molecular structure of fecal matter so how can you be sure it's shite?"
One person being ranked inappropriately does not undermine the entirety of the system.
Much like all systems, anomalies occur. Get on a lucky winning streak and your rank goes up. Tank for 3-4 months and your rank goes down. This is true for all systems. The players who are active will always go up in any and all ranked systems.
On April 19 2013 18:57 TheBB wrote: SlixSC:The reason ForGG is on #4 is not because "the formula is inaccurate" (although it could be). It's because the data is segregated. I could tweak the system so that ForGG is not on fourth place, but that might not satisfy you because your criticism is so extremely specific that it's totally useless.
The problem is that you're not being constructive. You're only pointing out problems without offering solutions, and I'm tired of it.
Not sure if it is done already. But have you considered adding different factors of point swings depending on the enviroment the matches are played (tiers of tournaments) like I stated about MTG before?
Yep. When I have it ready you will see offline matches having about twice the effect. The reason I'm holding back is because I'm planning on making one change and then finalize the whole thing, so that people can rely on them a bit more long term (instead of like now when they've been changing once a month).
I've also considered other factors, but there are always something holding me back :p. But like I said, I'm open to being persuaded.
doesnt forgg being #4 mean that he has good chance of beating other players in his range? for example if forgg comes up on liquibet against leenock, the prediction will say forgg will win x%?
if i'm understanding correctly, the ranks are the result of calculations and those calculations also have predictability powers.
i think its quiet ignorant to claim "my list is better than yours, and i say forgg is not even top 10" solely based on his opinion meanwhile this ranking is a result of a system regardless of anyone likes it or not, and it works according to TheBB.
On April 20 2013 03:23 jinorazi wrote: doesnt forgg being #4 mean that he has good chance of beating other players in his range? for example if forgg comes up on liquibet against leenock, the prediction will say forgg will win x%?
if i'm understanding correctly, the ranks are the result of calculations and those calculations also have predictability powers.
i think its quiet ignorant to claim "my list is better than yours, and i say forgg is not even top 10" solely based on his opinion meanwhile this ranking is a result of a system regardless of anyone likes it or not, and it works according to TheBB.
too many people getting butthurt for no reason :/
The internet would be a lot less interesting if people weren't butt hurt all the time. I'd have to get off the PC and like, live life and shit.
I don't think the system is bad. It will produce good results in time.
But the big mistake was trying to tell us something about the game, and basing it on questionable top 10. You can't show people, who in fact don't care about math, results which have obvious flaws. They don't care about math or system, but they will eat you, if you show them results like that.
There may be problem with data, but the biggest problem was creating this thread. If you waited a little while and showed some better output, you would get a lot better advertisement. You don't want people to remember "the system where forGG is 4th best in the world".
All math is great and all, but in the end you need experienced mind to look at results. At least that's how it works in physics.
From reading part of the thread, it seems that your algorithm takes into account other players that a certain player has played and 'takes' points from a player for winning.
For example, if forGG has been 'french-plowing', this means his high rating shows that his skill is far above that of his opponents.
Suppose the region he's playing in is isolated, i.e. they rarely play NA or KR players, doesn't this mean there is a low correlation between forGG's rating and other KR player ratings?
It seems to say that it is much more accurate if you compare players within a region, but has more uncertainty when compared to other regions. This means that if forGG suddenly moves to Korea, his rating would take some time to stabilize again. Of course this is on the assumption that regions are isolated.
On April 20 2013 04:21 Tuczniak wrote: I don't think the system is bad. It will produce good results in time.
But the big mistake was trying to tell us something about the game, and basing it on questionable top 10. You can't show people, who in fact don't care about math, results which have obvious flaws. They don't care about math or system, but they will eat you, if you show them results like that.
There may be problem with data, but the biggest problem was creating this thread. If you waited a little while and showed some better output, you would get a lot better advertisement. You don't want people to remember "the system where forGG is 4th best in the world".
All math is great and all, but in the end you need experienced mind to look at results. At least that's how it works in physics.
Agreed, we hoped to show off the potential of the site and ofc stats doesn't equal the truth. It is just a basis for discussion, I guess most people aren't prepared to think for themselves.
On April 20 2013 04:39 OrgCom wrote: From reading part of the thread, it seems that your algorithm takes into account other players that a certain player has played and 'takes' points from a player for winning.
For example, if forGG has been 'french-plowing', this means his high rating shows that his skill is far above that of his opponents.
Suppose the region he's playing in is isolated, i.e. they rarely play NA or KR players, doesn't this mean there is a low correlation between forGG's rating and other KR player ratings?
It seems to say that it is much more accurate if you compare players within a region, but has more uncertainty when compared to other regions. This means that if forGG suddenly moves to Korea, his rating would take some time to stabilize again. Of course this is on the assumption that regions are isolated.
Over time trends will form--those trends are what are fun to watch.
People get excited from regional success and cross their fingers hoping it translates to cross regional success.
Can the people criticizing it whilst offering absolutely nothing constructive please remember:
The accuracy of this ranking supercedes anything produced so far in SC2 and maybe in any esport.
Using this ranking would have you absolutely dominate a randomly selected individual in liquibets on average, even if you only selected that individual from a group of hardcore fans/casters and players.
So it DOES have merit, irrefutably. I keep seeing the same stuff "oh its numbers and maths blah blah but its not applicable to real life play" etc etc. Well actually, it is. It has very good predictive ability which only increases with time. You people have exactly the same kind of dinosaur way of thinking that riddles professional sport. Think of it like this: This is moneyball and you are the old dinosaur scouts who judge based one 'intuition'.
The ranking list is vastly superior to anything you as an individual could come up with in accuracy even if you disagree with it.
And no they are not ignoring your constructive criticism you don't understand what constructive means (to that absolutely clueless Austrian guy especially who was posting on the last page, my god what a low IQ cretin. Prediction he is no more than 15 years old and does not do well at maths at school).
On April 20 2013 05:53 _SpiRaL_ wrote: Can the people criticizing it whilst offering absolutely nothing constructive please remember:
The accuracy of this ranking supercedes anything produced so far in SC2 and maybe in any esport.
Using this ranking would have you absolutely dominate a randomly selected individual in liquibets on average, even if you only selected that individual from a group of hardcore fans/casters and players.
So it DOES have merit, irrefutably. I keep seeing the same stuff "oh its numbers and maths blah blah but its not applicable to real life play" etc etc. Well actually, it is. It has very good predictive ability which only increases with time. You people have exactly the same kind of dinosaur way of thinking that riddles professional sport. Think of it like this: This is moneyball and you are the old dinosaur scouts who judge based one 'intuition'.
The ranking list is vastly superior to anything you as an individual could come up with in accuracy even if you disagree with it.
And no they are not ignoring your constructive criticism you don't understand what constructive means (to that absolutely clueless Austrian guy especially who was posting on the last page, my god what a low IQ cretin. Prediction he is no more than 15 years old and does not do well at maths at school).
From what I remember the system use something similar to true skill, so it is not "flawed", just it will take some time to converge, especially with the arrival of a lot of new players (the kespa player). What would make it converge a lot sooner is to use something like true skill through time, to incorporate futre results into the present ranking.
On April 20 2013 06:21 gondolin wrote: From what I remember the system use something similar to true skill, so it is not "flawed", just it will take some time to converge, especially with the arrival of a lot of new players (the kespa player). What would make it converge a lot sooner is to use something like true skill through time, to incorporate futre results into the present ranking.
So guessing the skill in the future?
Obviously we can always use forward filthering backwards smoothing once games are played and make the older rankings better, but that's not really doing us any good with the newest ranking.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
On April 20 2013 16:03 Big-t wrote: Who is Bunny? Top 10 foreigners and I have never heard of him????
Last year he got 3rd at WCS Denmark.
Goes to show how much you know about the foreign scene really..
News flash, a huge majority of StarCraft fans do not care about the results of WCS Denmark. Also that bracket doesn't look intimidating in the slightest.
On April 20 2013 16:03 Big-t wrote: Who is Bunny? Top 10 foreigners and I have never heard of him????
Last year he got 3rd at WCS Denmark.
Goes to show how much you know about the foreign scene really..
News flash, a huge majority of StarCraft fans do not care about the results of WCS Denmark. Also that bracket doesn't look intimidating in the slightest.
If they don't care, they would do well to not post with such unwarranted self importance, since their opinion is uneducated. ^^
Plus, he qualified for the WCS EU Premier with wins over LiveZerg, Welmu and MVP.finale, and also won a showmatch against monchi 4-1. I'd say he's fairly strong as of late.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", he replied to a post which was referencing win rates and he too was talking about win rates.
If you guys are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
But then to go as far as to say that my posts are "stupid" when you yourself seem to fail to understand the context of the discussion is actually laughable, truly laughable.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
On April 20 2013 16:26 figq wrote: Don't you dare question the mighty relevance of getting 3rd in WCS Denmark. The man is a legend.
While I do understand the comical aspect of this. Bunny's points mainly stem from him smacking around the Danish scene and qualifying for WCS EU playing very strong players.
Our list rewards players who go on hot streaks with points. That is the basic premise of the system.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
And the guy also includes stuff like Zotac Final as well, is that considered "Pro enough"?
Zotac Finals are professional players only.
Hyun, beastyqt, Arthur, Nerchio, revival, tarra, happy and bly
I think that's pro enough
Our biggest "non-pro" is the DSCL qualifiers I think. But all players there have sub 1000 points.
Last month mainly consist of GSL, GSTL, Pro-League, WCS qualifiers, Acer Teamstory Cup, Gigabyte cups, Yegalisk Master something, Various Ro8/4's from G4SC2 cups etc.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
And the guy also includes stuff like Zotac Final as well, is that considered "Pro enough"?
Zotac Finals are professional players only.
Hyun, beastyqt, Arthur, Nerchio, revival, tarra, happy and bly
I think that's pro enough
Should we disconsider DreamHack then, since they tend to have a few of amateur/semi-pro players in the early group stages?
Do you see how arbitrary this is? What if there are are "amateur" players who aren't on a proteam and don't practice 6 or 8 hrs a day but can still manage to beat a professional in some small daily/weekly cup? Should that not count? And then there are countless of players who *are* on teams, not big teams granted, but pro teams nonetheless, but they don't win tournaments. Should their matches not count either?
And then there's the fallacy of "I'm not familiar with this player therefore he's not a pro or simply terrible"... even when they've actually won stuff. ._.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
And the guy also includes stuff like Zotac Final as well, is that considered "Pro enough"?
Zotac Finals are professional players only.
Hyun, beastyqt, Arthur, Nerchio, revival, tarra, happy and bly
I think that's pro enough
Should we disconsider DreamHack then, since they tend to have a few of amateur/semi-pro players in the early group stages?
Do you see how arbitrary this is? What if there are are "amateur" players who aren't on a proteam and don't practice 6 or 8 hrs a day but can still manage to beat a professional in some small daily/weekly cup? Should that not count? And then there are countless of players who *are* on teams, not big teams granted, but pro teams nonetheless, but they don't win tournaments. Should their matches not count either?
And then there's the fallacy of "I'm not familiar with this player therefore he's not a pro or simply terrible"... even when they've actually won stuff. ._.
Sure it's arbitrary. But this will always be the case. Ignoring diamond league players in your balance statistics isn't any less arbitrary than ignoring players you don't consider pro. You have to start at some point and if the early stages of dreamhack tend to have a significant amount of non-pro players then I think you would be justified in ignoring the first group stage for example.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
And the guy also includes stuff like Zotac Final as well, is that considered "Pro enough"?
Zotac Finals are professional players only.
Hyun, beastyqt, Arthur, Nerchio, revival, tarra, happy and bly
I think that's pro enough
Our biggest "non-pro" is the DSCL qualifiers I think. But all players there have sub 1000 points.
Last month mainly consist of GSL, GSTL, Pro-League, WCS qualifiers, Acer Teamstory Cup, Gigabyte cups, Yegalisk Master something, Various Ro8/4's from G4SC2 cups etc.
Which is fair enough. I still think you should ignore showmatches though, for reasons I already mentioned.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
And the guy also includes stuff like Zotac Final as well, is that considered "Pro enough"?
Zotac Finals are professional players only.
Hyun, beastyqt, Arthur, Nerchio, revival, tarra, happy and bly
I think that's pro enough
Should we disconsider DreamHack then, since they tend to have a few of amateur/semi-pro players in the early group stages?
Do you see how arbitrary this is? What if there are are "amateur" players who aren't on a proteam and don't practice 6 or 8 hrs a day but can still manage to beat a professional in some small daily/weekly cup? Should that not count? And then there are countless of players who *are* on teams, not big teams granted, but pro teams nonetheless, but they don't win tournaments. Should their matches not count either?
And then there's the fallacy of "I'm not familiar with this player therefore he's not a pro or simply terrible"... even when they've actually won stuff. ._.
Sure it's arbitrary. But this will always be the case. Ignoring diamond league players in your balance statistics isn't any less arbitrary than ignoring players you don't consider pro. You have to start at some point and if the early stages of dreamhack tend to have a significant amount of non-pro players then I think you would be justified in ignoring the first group stage for example.
You are including show matches and a significant amount of amateur games. Including games that have nothing to do with pro play is kind of defeating the whole purpose of a balance graph. Why not take it to the extreme and also include bronze matches and practise matches? You are including amateur matches and showmatches after all so it doesn't seem that far off.
Wat. This isn't a balance graph. It's a system that attempts to predict the outcome of a match with some degree of accuracy, based upon a player rating, which is calculated from their match history. If you seriously think showmatches between pro players are not remotely representative of their skill level at that given point in time, then...
I suggest you stop posting in this thread, you're doing nothing but making a fool out of yourself with stupid claim upon stupid claim none of which would happen if you had even read the information provided in the OP or in the site's FAQ.
He was talking about win rates. So yes, he thinks including amateur matches and showmatches has statistical relevance when talking about win rates. Which is bullshit.
First of all, show matches are completely different to competitive matches, players will more likely try to play longer games (it's a show match) which will inevitably skew the outcome of the game one way or another.
Amateur matches are amateur matches, they tell you exactly bupkis about the current state of balance.
I honestly don't see how my claims "are stupid", you can't make an argument and as soon as someone calls you out for it go "that's not what he/I actually meant, read the FAQ".
If you are using language incorrectly or ambigously you can't just assume that people will interpret your posts in the most charitable way possible.
I do agree. However showmatches with money on the line do hold relevance to most people. If you want "by event" balance stats (although they have very few matches) we do offer those on every event in our DB.
And the guy also includes stuff like Zotac Final as well, is that considered "Pro enough"?
Zotac Finals are professional players only.
Hyun, beastyqt, Arthur, Nerchio, revival, tarra, happy and bly
I think that's pro enough
Our biggest "non-pro" is the DSCL qualifiers I think. But all players there have sub 1000 points.
Last month mainly consist of GSL, GSTL, Pro-League, WCS qualifiers, Acer Teamstory Cup, Gigabyte cups, Yegalisk Master something, Various Ro8/4's from G4SC2 cups etc.
Which is fair enough. I still think you should ignore showmatches though, for reasons I already mentioned.
I disagree with your assesment of showmatches which are sponsored, thus having money on the line are, since playing for money is the bread and butter of most proplayers.
Our policy is to include rounds where we can be reasonably sure of the identities of most of the players involved, and to include players who we can be reasonably sure will not keep playing in a bubble (that is, they will either keep interacting with the greater pool, or they will play only a few matches and then vanish in obscurity). There's a fair amount of leeway still, of course.
For example, for the WCS EU qualifiers, the Ro64 tended to have maybe 70-80% well known players, so that's where we started adding.
A quick look indicates that showmatches make up 0.44% of the matches in the database (250). That seems like an unlikely source of error IMO.
On April 20 2013 06:21 gondolin wrote: From what I remember the system use something similar to true skill, so it is not "flawed", just it will take some time to converge, especially with the arrival of a lot of new players (the kespa player). What would make it converge a lot sooner is to use something like true skill through time, to incorporate futre results into the present ranking.
So guessing the skill in the future?
Obviously we can always use forward filthering backwards smoothing once games are played and make the older rankings better, but that's not really doing us any good with the newest ranking.
Well yes You could filter backwards to make the older rankings better and use these older rankings to improve the current rankings.
Imagine that two new players "Life" and "Flash" appears out of nowhere. Their first match is a BO9 between them where Flash win 5-4. The system can determine they are pretty close in skill, but since they are new players their rating does not change much. Then a week latter, Life goes on to win the GSL without dropping a single map, while Flash does not play at all. Then in the current system Life will have a lot of points but Flash rating will stay the same. So the problem is that the system determined that Life is really good *after* his BO9 with Flash. If they had played the BO9 after the GSL, the system would have determined that Flash also is very good.
But what you can do is now that you know that Life is very good, since he was probably also very good last week is to update the current rating of Flash. You can look at True Skill through time or Whole history rating to see how to incorporate past results retroactively into present ratings in a converging manner.
One argument against this system is that it would change past rating, but what you can do is only change these past ratings internally to have better current ratings. (One drawback of this system would still be that the rating of a player will change even if he does not play any match, but that's exactly the point!)
I have a few suggestions such as new balance stats, a RacevRace win% graph that only counts games between the top X% of pros, say 33%, and another that disconsiders games between players with too much rating discrepancy if the player with more rating wins.
As someone that watches a lot of SC2 the graph showing the ebb and flow of race balance since 2010 looks very accurate. Will be very interesting to see what the next plot looks like. Whether Z and P have started to work out Terran or whether Terran makes even more use of its new units
EPT
Life 1921
Polt 1837
PartinG 1798
![[image loading]](http://i.imgur.com/1hcSTvA.png)
. That would have the biggest effect. Maybe Gumiho.![[image loading]](http://static.aligulac.com/predict.png)
