|
On April 19 2013 17:41 Grovbolle wrote:Show nested quote +On April 19 2013 17:35 SlixSC wrote:On April 19 2013 17:19 Grovbolle wrote:On April 19 2013 17:11 SlixSC wrote:On April 19 2013 17:07 Grovbolle wrote: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 Show nested quote +How predictive is the system???? By my understanding, very. ![[image loading]](http://static.aligulac.com/predict.png) 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?"
|
Denmark3807 Posts
On April 19 2013 18:38 SlixSC wrote:Show nested quote +On April 19 2013 17:41 Grovbolle wrote:On April 19 2013 17:35 SlixSC wrote:On April 19 2013 17:19 Grovbolle wrote:On April 19 2013 17:11 SlixSC wrote:On April 19 2013 17:07 Grovbolle wrote: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.
|
Denmark3807 Posts
On April 19 2013 18:52 Scarecrow wrote:Show nested quote +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.
And thanks for calling me an asshole.
I'll show myself out.
|
Denmark3807 Posts
On April 19 2013 19:03 SlixSC wrote:Show nested quote +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.
http://www.teamliquid.net/forum/viewmessage.php?topic_id=408365¤tpage=5#87
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.
|
protoss are fucked, as always 
|
I really like the page design. So elegant and simple, as it should be.
|
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.
|
So turns out ForGG might not be half bad after all. :p
|
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?
|
Denmark3807 Posts
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 18:38 SlixSC wrote:Show nested quote +On April 19 2013 17:41 Grovbolle wrote:On April 19 2013 17:35 SlixSC wrote:On April 19 2013 17:19 Grovbolle wrote:On April 19 2013 17:11 SlixSC wrote:On April 19 2013 17:07 Grovbolle wrote: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 20 2013 03:08 Usagi wrote:Show nested quote +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.
too many people getting butthurt for no reason :/
|
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.
|
|
|
|
|
|
|
|
EPT
