2. Byun
3. TY
4. Solar
5. Stats
6. Patience
7. Zest
8. Neeb
9. Dear
10. Nerchio
11. TRUE
12. Snute
13. Showtime
14. Elazer
15. Violet
16. Ptitdrogo
Forum Index > SC2 General |
Azhrak
Finland1192 Posts
2. Byun 3. TY 4. Solar 5. Stats 6. Patience 7. Zest 8. Neeb 9. Dear 10. Nerchio 11. TRUE 12. Snute 13. Showtime 14. Elazer 15. Violet 16. Ptitdrogo | ||
Hider
Denmark9342 Posts
On October 25 2016 22:50 AbouSV wrote: Articles (whether TL or not) are not meant to be 'Statistic analysis master degree final exam', but they can give any of the following: background, recent form, general information, humour, hype, and, yes, biais (the subjective one, not the mathematical one). Someone good at stats can do the math himself. If not, the details of such an abstract article would just be pointless to read anyway. It matters for me. Because now people will probably think Neeb choked - as in mental weakness - when he "unexpectedly" goes out in quarter finals. Also, if you are not taking data seriously, then I would like to see more focus on the playstyles of the players. What makes them unique? What are their strenghts/weakness's, and then highlight a VOD that shows how the player typically plays. | ||
Elentos
55456 Posts
On October 25 2016 23:21 Hider wrote: Show nested quote + On October 25 2016 22:50 AbouSV wrote: Articles (whether TL or not) are not meant to be 'Statistic analysis master degree final exam', but they can give any of the following: background, recent form, general information, humour, hype, and, yes, biais (the subjective one, not the mathematical one). Someone good at stats can do the math himself. If not, the details of such an abstract article would just be pointless to read anyway. It matters for me. Because now people will probably think Neeb choked - as in mental weakness - when he "unexpectedly" goes out in quarter finals. Well then the course of action is obvious, Hider. You go there and rig the whole thing so that Neeb and Dark play in the quarter finals. #2 going out against #1 isn't "unexpected" or an "upset". Also prepare a time machine in case Dark messes up and loses or doesn't win convincingly enough for it to not be a "choke". | ||
Fango
United Kingdom8987 Posts
On October 25 2016 21:40 Hider wrote: Show nested quote + You can't decide purely based of winrates. Nerchio and Byun have the most inflated winrates because 90% of games they play are against B/C tier players I don't, hence the criteria was better than Kelazhur. Just look through the data yourself. Show nested quote + What top level guys has Nerchio beat in his entire career? Classic? Is that really it? Well given how highly you seem to value Neeb, it should be pointed out that Nerchio won 6-3 against Neeb in their most recent encounter. But your methodology is just flawed. Your just nitpicking very specific results instead of just taking a broader view of all of the matches both players have played, and then weighted them based on opponents skill level and date. When you have a large sample size, you don't sit there and overanalyze 1-2 games. No (!) - you let data speak for it self. And Neeb's recent overall results have been mediocre - hence why his ELO ranking has gone down. Except those "1-2 games" are only thing to go off as far as nerchio vs koreans. Nerch has faced Koreans in this kind of tournament once (kespa cup). And he didnt cut it. Ontop of that he regularly underperforms against other foreigners even though he's meant to be the "foreigner king". Please explain how he can lose to lilbow, scarlett, marinelord, and snute in bo5 series, and then you people say he still beats TY, Byun, dark etc be cause of his winrates. Just...how??? On October 25 2016 21:36 Elentos wrote: Show nested quote + On October 25 2016 21:21 Fango wrote: On October 25 2016 20:29 Hider wrote: On October 25 2016 20:14 Fango wrote: On October 25 2016 20:03 Hider wrote: The fact that they played their best matchups does not diminish their achievement. On the other hand, I think Nerchio on a good day can beat everyone (he has a chance vs the top 4 players - still a small underddog ofc). I dunno. Patience and maru beat him very hard. His best matchup against koreans is ZvP. Thats the only matchup I could favour him in Eh, he is 2-2 vs Maru over the last 2 months. 1-0 vs Byun. And if you take a wider look at his match history, you generally see he has very good results against some of the better terrans out there. I find it problematic when people make predictions only based on the very few games they have watched of him rather than just skimming through his match history. http://aligulac.com/players/26-Nerchio/results/ On the other hand Neeb can barely beat Kelazhur and losses to Scarlett. Nerchio is like 35-3 over the last 2 months against terrans of equal or higher skill than Kelazhur. Maru destroyed him in the series they played, that third map made the game look broken The first map was 28 minutes long and Maru was behind for like 23 of those 28 minutes, he barely won because Nerchio messed up. He then lost the next map. And then he had that game 3 which was the only one where he won convincingly. That is so far off from any definition of "He destroyed him" I could agree with. The first map Nerchio was completely clueless on what to do against Maru army (he even admitted this on TL after the match). Maru was gonna win no matter how long the game was. Game 2 was very close sure and Nerch defended very well, I'm not quite sure what Maru was trying to do (maybe he just really wanted to play map 3 idk). But game 3 was an utter steamroll, Nerch never had a good position all game long. I'd say thats a one sided series. Nerch only looked like he could have won in game 2 and even then it was close. (In shoutcraft it was a 100% clear win for nerchio to put a perspective on it). In a similar way to when Zest vs Maru in season 1 was 3-0, but was incredibly close every game. Results don't always show how close a series really is. | ||
Hider
Denmark9342 Posts
Except those "1-2 games" are only thing to go off as far as nerchio vs koreans. Nerch has faced Koreans in this kind of tournament once (kespa cup). You narrow it down to a very specific set of circumstances, and only under those circumstances does the results matter. For some reason there exist this myth that online results should be ignored, even though the correlation is insanely high between offline and online results. And for some reason, results against top foreigners should be ignored because they cannot be compared to Koreans in any way sort of shape. Even though Neeb's results were mostly against non tier 1 koreans. According to your methodology, if someone has a 90% win/rate against good foreigners that data information should be completely ignored when estimating how they will perform against koreans. But in proper data analysis, you look at every single information, and then you weight them accordingly in order to maximize the predictive power. | ||
![]()
Poopi
France12759 Posts
Like putting Neeb at #2 is worse delusional hyping than NA LoL teams have ever been hyped, even tho they are overhyped/overrated every year. When you barely beat a brasilian player in PvT, and lose against Scarlett, win a korean tourney relying almost solely on PvP, perform worse all year long than the best foreigner, you are now considered #2 for Blizzcon? Poor ByuN shits on Neeb all day everyday in practice in order to beat equal/better protosses, is great in every matchup, shits on everyone on Aligulac and in online tourneys, but hey Neeb is good at PvP so you have a worse Power Ranking ![]() It's comical in a bad way. | ||
Drfilip
Sweden590 Posts
![]() 2. ![]() 3. ![]() 4. ![]() 5. ![]() 6. ![]() 7. ![]() 8. ![]() 9. ![]() 10. ![]() 11. ![]() 12. ![]() 13. ![]() 14. ![]() 15. ![]() 16. ![]() | ||
Fango
United Kingdom8987 Posts
On October 26 2016 00:54 Hider wrote: Show nested quote + Except those "1-2 games" are only thing to go off as far as nerchio vs koreans. Nerch has faced Koreans in this kind of tournament once (kespa cup). You narrow it down to a very specific set of circumstances, and only under those circumstances does the results matter. For some reason there exist this myth that online results should be ignored, even though the correlation is insanely high between offline and online results. And for some reason, results against top foreigners should be ignored because they cannot be compared to Koreans in any way sort of shape. Even though Neeb's results were mostly against non tier 1 koreans. According to your methodology, if someone has a 90% win/rate against good foreigners that data information should be completely ignored when estimating how they will perform against koreans. But in proper data analysis, you look at every single information, and then you weight them accordingly in order to maximize the predictive power. I am NOT only using his Kespa cup results you've missed my other point. Nerchios results are not good offline vs top foreigners or koreans. I repeat what I said before. Explain to me how, if he loses to Lilbow, Scarlett, Snute, or MLord in a basic bo5 with a few days prep, will he ever take on top starleague players? Because to me that makes no sense. aligulac %s mean NOTHING when you can't take wins in tourneys. Offline results ARE completely different. SOMETIMES they can be a sign of skill (i.e. Byun). sometimes not (i.e everyone who isn't Byun). The thing is a lot of the best players never do them. If Dark or Maru or Dear or TY actually did many online cups their winrates would be 20-30% higher than they are. Heck, Dark's winrates are not even that high despite him being the clear best player of the year with 4 finals and 2 trophies. | ||
WonnaPlay
Netherlands912 Posts
![]() 2. ![]() 3. ![]() 4. ![]() 5. ![]() 6. ![]() 7. ![]() 8. ![]() 9. ![]() 10. ![]() 11. ![]() 12. ![]() 13. ![]() 14. ![]() 15. ![]() 16. ![]() | ||
Jisira
470 Posts
![]() 2. ![]() 3. ![]() 4. ![]() 5. ![]() 6. ![]() 7. ![]() 8. ![]() 9. ![]() 10. ![]() 11. ![]() 12. ![]() 13. ![]() 14. ![]() 15. ![]() 16. ![]() | ||
Hider
Denmark9342 Posts
Offline results ARE completely different. SOMETIMES they can be a sign of skill (i.e. Byun). sometimes not (i.e everyone who isn't Byun) And people said the same thing about Neeb. Actually try this: Make a statistical model and then completely ignore online results when attempting to predict offline results, and compare the accuracy of that model to a model that actually takes into account online results. Also, please sign up to a bookmaker and place your money where your mouth is. | ||
Wrathsc2
United States2025 Posts
2. Solar 3. Byun 4. Stats 5. TY 6. Zest 7. Neeb 8. Patience 9. Nerchio 10. Dear 11. True 12. Snute 13. Violet 14. Elazer 15. Showtime 16. PtitDrogo | ||
Fango
United Kingdom8987 Posts
On October 26 2016 03:16 Hider wrote: Show nested quote + Offline results ARE completely different. SOMETIMES they can be a sign of skill (i.e. Byun). sometimes not (i.e everyone who isn't Byun) And people said the same thing about Neeb. Actually try this: Make a statistical model and then completely ignore online results when attempting to predict offline results, and compare the accuracy of that model to a model that actually takes into account online results. Also, please sign up to a bookmaker and place your money where your mouth is. My lord you have tunnel vision. I didn't say not to take online cups into account, but they are NOT always indicative of performance. Simply looking at recent offline performacne is the only way to guess how a player is doing in tournaments right now. I'm sorry I will never use a statistical model to predict sc results because it doesn't take into account so many factors. Don't get me wrong percentages are cute to look at and can be useful. But anyone who has watched all the recent tournaments and followed the players for years will have more accurate predictions than a bot. Now Ima repeat myself for the 5th and hopefully final time- "If Nerchio cannot beat high level foriengers like Snute, Scarlett, Mlord, or even Lilbow in a straight up bo5, he will never be able to take out top starleague players" You are joking if you think he can because he has "good online winrates" haha you think if Dark or TY or Zest did as many online cups they wouldn't have even higher percentages??? Or have you only been around the scene this year and don't know the differences between levels of play. | ||
![]()
Poopi
France12759 Posts
On October 26 2016 04:02 Fango wrote: Show nested quote + On October 26 2016 03:16 Hider wrote: Offline results ARE completely different. SOMETIMES they can be a sign of skill (i.e. Byun). sometimes not (i.e everyone who isn't Byun) And people said the same thing about Neeb. Actually try this: Make a statistical model and then completely ignore online results when attempting to predict offline results, and compare the accuracy of that model to a model that actually takes into account online results. Also, please sign up to a bookmaker and place your money where your mouth is. My lord you have tunnel vision. I didn't say not to take online cups into account, but they are NOT always indicative of performance. Simply looking at recent offline performacne is the only way to guess how a player is doing in tournaments right now. I'm sorry I will never use a statistical model to predict sc results because it doesn't take into account so many factors. Don't get me wrong percentages are cute to look at and can be useful. But anyone who has watched all the recent tournaments and followed the players for years will have more accurate predictions than a bot. Now Ima repeat myself for the 5th and hopefully final time- "If Nerchio cannot beat high level foriengers like Snute, Scarlett, Mlord, or even Lilbow in a straight up bo5, he will never be able to take out top starleague players" You are joking if you think he can because he has "good online winrates" haha you think if Dark or TY or Zest did as many online cups they wouldn't have even higher percentages??? Or have you only been around the scene this year and don't know the differences between levels of play. By this metric Neeb lost to Scarlett and barely beat Kelazhur (afaik it was offline, it wouldn't change much if it was online tho ![]() Nerchio has beaten all the players you listed in bo5 afaik. Nerchio 4-0 MLorD DH Valencia 2016. (offline) Nerchio 3-1 Scarlett WCS Spring 2016. (offline) Nerchio 4-2 Lilbow Acer Pro Challenge (online) Nerchio 3-2 Snute many times (online or offline) Nerchio is the best foreigner by far in LotV and he has the most chances of doing well among foreigners. | ||
Fango
United Kingdom8987 Posts
On October 26 2016 04:13 Poopi wrote: Show nested quote + On October 26 2016 04:02 Fango wrote: On October 26 2016 03:16 Hider wrote: Offline results ARE completely different. SOMETIMES they can be a sign of skill (i.e. Byun). sometimes not (i.e everyone who isn't Byun) And people said the same thing about Neeb. Actually try this: Make a statistical model and then completely ignore online results when attempting to predict offline results, and compare the accuracy of that model to a model that actually takes into account online results. Also, please sign up to a bookmaker and place your money where your mouth is. My lord you have tunnel vision. I didn't say not to take online cups into account, but they are NOT always indicative of performance. Simply looking at recent offline performacne is the only way to guess how a player is doing in tournaments right now. I'm sorry I will never use a statistical model to predict sc results because it doesn't take into account so many factors. Don't get me wrong percentages are cute to look at and can be useful. But anyone who has watched all the recent tournaments and followed the players for years will have more accurate predictions than a bot. Now Ima repeat myself for the 5th and hopefully final time- "If Nerchio cannot beat high level foriengers like Snute, Scarlett, Mlord, or even Lilbow in a straight up bo5, he will never be able to take out top starleague players" You are joking if you think he can because he has "good online winrates" haha you think if Dark or TY or Zest did as many online cups they wouldn't have even higher percentages??? Or have you only been around the scene this year and don't know the differences between levels of play. By this metric Neeb lost to Scarlett and barely beat Kelazhur (afaik it was offline, it wouldn't change much if it was online tho ![]() Nerchio has beaten all the players you listed in bo5 afaik. Nerchio 4-0 MLorD DH Valencia 2016. (offline) Nerchio 3-1 Scarlett WCS Spring 2016. (offline) Nerchio 4-2 Lilbow Acer Pro Challenge (online) Nerchio 3-2 Snute many times (online or offline) Nerchio is the best foreigner by far in LotV and he has the most chances of doing well among foreigners. Very true but he has also lost to all of them in bo5 more recently offline as fair as I'm aware. I would agree him Snute and Neeb are the best foreigners for years. It's just I can't see him having a chance at beating a top 5 Kespa player when he still can't consistantly beat top 10 foreigners. It's like Lilbow syndrome again. That man was so hyped last blizzcon due to his results in WCS europe (or whatever it was) but was still a full tier below when it came to facing Life. | ||
Hider
Denmark9342 Posts
My lord you have tunnel vision. I didn't say not to take online cups into account, but they are NOT always indicative of performance. No you actually directly saying to ignore online performances: . Simply looking at recent offline performacne is the only way to guess how a player is doing in tournaments right now. So recent online performances has no impact on how well we can expect the players to do in the next offline tournament. It seems like you don't even understand the implications of what you are actually writing. "If Nerchio cannot beat high level foriengers like Snute, Scarlett, Mlord, or even Lilbow in a straight up bo5, he will never be able to take out top starleague players" What? Obvivously he can. What the hell are you on? You are joking if you think he can because he has "good online winrates" haha you think if Dark or TY or Zest did as many online cups they wouldn't have even higher percentages??? Or have you only been around the scene this year and don't know the differences between levels of play. No idea what you are talking about. But you seem like someone who just spew absolutely random shit. Or have you only been around the scene this year and don't know the differences between levels of play. Let's check your profile: Joined TL.net Saturday, 2nd of July 2016 Now let's check mine: Joined TL.net Wednesday, 5th of May 2010 So pretty much this confirms your just spewing random bullshit based on nothing. No need to talk to you further. Have a nice life bye. | ||
NinjaToss
Austria1383 Posts
2. TY 3. Byun 4. Solar 5. Stats 6. Neeb 7. Snute 8. Patience 9. Nerchio 10. Dear 11. ShoWTimE 12. TRUE 13. Zest 14. Violet 15. Elazer 16. Ptitdrogo | ||
![]()
Soularion
Canada2764 Posts
![]() | ||
Hider
Denmark9342 Posts
On October 26 2016 05:43 Soularion wrote: We'll see how things go when it comes for Blizzcon. Should invent some kind of a system for showing which PR is best ![]() PR doesn't take into account schedule difficulty so we can't do that. | ||
FrkFrJss
Canada1205 Posts
2. Byun 3. Stats 4. TY 5. Solar 6. Neeb 7. Zest 8. Patience 9. Dear 10. Nerchio 11. Snute 12. TRUE 13. Showtime 14. Elazer 15. Violet 16. Ptitdrogo It's so tough...I feel like I should rank by matchup, because I would favour Neeb over any other Protoss, but his other matchups haven't been as good. And a lot of Koreans haven't been performing as well, but they might still be better than the best foreigners. | ||
| ||
![]() StarCraft 2 StarCraft: Brood War Stormgate Dota 2 League of Legends Counter-Strike Super Smash Bros Other Games Organizations
StarCraft 2 • practicex StarCraft: Brood War![]() • v1n1z1o ![]() • AfreecaTV YouTube • intothetv ![]() • Kozan • IndyKCrew ![]() • sooper7s • LaughNgamezSOOP • Migwel ![]() League of Legends |
Wardi Open
Monday Night Weeklies
PiGosaur Monday
OSC
Code For Giants Cup
The PondCast
Replay Cast
SC Evo Complete
Classic vs uThermal
SOOP StarCraft League
CranKy Ducklings
[ Show More ] [BSL 2025] Weekly
SOOP StarCraft League
Sparkling Tuna Cup
uThermal 2v2 Circuit
|
|