|
On January 10 2014 03:50 nojok wrote:Show nested quote +On January 10 2014 03:28 Xiphos wrote:On January 10 2014 03:26 nojok wrote:
TL's BW articles are always top notch. I don't even follow BW but they're good and I appreciate the links to games worth watching and all fun Korean stuff. Maybe that's something you think of doing  I will give it a shot I think now that the Korean scene seems to improve recently. I was disappointed overall by the last matches I watch compared to the OSL/MSL era, though it was mostly foreigner matches.
Watch game 3 of the first series of Killer vs Mong that occurred just a couple of hours ago. You won't regret it.
1:20:00
http://www.twitch.tv/caucasianasian1/c/3519921
|
On January 09 2014 15:29 miercat wrote:Show nested quote +On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%)
I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success.
Soulkey
Round of 36(1): Bacchus OSL
Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL
ZerO
Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL
Round of 16(1): Nate MSL,
Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2
Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL
Runner-Up(2): ABC Mart MSL, Blizzcon 2009
Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament.
|
On January 10 2014 04:11 Xiphos wrote:Show nested quote +On January 10 2014 03:50 nojok wrote:On January 10 2014 03:28 Xiphos wrote:On January 10 2014 03:26 nojok wrote:
TL's BW articles are always top notch. I don't even follow BW but they're good and I appreciate the links to games worth watching and all fun Korean stuff. Maybe that's something you think of doing I will give it a shot I think now that the Korean scene seems to improve recently. I was disappointed overall by the last matches I watch compared to the OSL/MSL era, though it was mostly foreigner matches. Watch game 3 of the first series of Killer vs Mong that occurred just a couple of hours ago. You won't regret it. 1:20:00 http://www.twitch.tv/caucasianasian1/c/3519921
Thank you, nice one indeed.
+ Show Spoiler +Why so few vessels!? And what a pullover!
|
On January 10 2014 04:44 lemmata wrote:Show nested quote +On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament.
I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case.
And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level.
Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things...
|
Bisu brings all the fangirls back to SSL! (much needed, complete sausage fest)
|
On January 10 2014 05:46 miercat wrote:Show nested quote +On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things...
Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong.
Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO.
Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. It's not an unrealistic hypothesis in terms of the small numbers (<1000) involved. EDIT: I guess I never compared Soulkey to Sea. I thought I had.
|
+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW.
Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on.
Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%).
I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question.
|
On January 10 2014 07:37 miercat wrote:+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW. Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on. Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%). I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question.
Let's set aside the quality of play for a second, because that can be subjective. I don't think you have any reason to be disturbed. I do not think that your excellent memory is incorrect. But aside from the results, your perception of them are subjective even if they are correct. I won't even disagree that ZvT is not ZerO's best match up. ZerO was known more for his ZvP.
For both players, their games with Fantasy makes up a large portion of their ZvT record (14 games for ZerO and 7 games for Soulkey). For Soulkey in particular, his dominance against Fantasy is one of the first things that are brought up when people discuss his predator-prey relationships with other players.
ZerO vs Terran in standard leagues: 57W 51L (52.78%)
ZerO vs Non-Fantasy Terran in standard leagues: 52W 42L (55.3%)
Soulkey vs Terran in standard leagues: 36W 31L (52.73%)
Soulkey vs Non-Fantasy Terran in standard leagues: 30W 30L (50%)
As far as the Korean comments, I am Korean so I am just speaking about discussions with other Koreans plus my perception of how each player was pumped up to the viewer by the Korean commentators (rather than a direct comparison).
Finally, speaking as a professional mathematician, I think that it is strange for 1) these number of games to be considered a large sample for statistical evaluation and 2) for a 5~7% difference in winning percentages to be considered meaningful over 100 games. However, this is also why I am willing to accept that I could be wrong and Soulkey could have been just unlucky in individual leagues.
EDIT: One area of disagreement: I don't think it is fair to describe ZerO's results in individual leagues as merely decent.
PS: I will say that I am a little biased because ZerO was my favorite Zerg at the end of BW. Even though Jaedong was obviously the best Zerg by far, I just liked ZerO more for reasons I can't really explain.
EDIT: "End of BW" obviously never happened since we're here! Man, that makes me feel good.
|
On January 10 2014 08:24 lemmata wrote:Show nested quote +On January 10 2014 07:37 miercat wrote:+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW. Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on. Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%). I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question. Let's set aside the quality of play for a second, because that can be subjective. I don't think you have any reason to be disturbed. I do not think that your excellent memory is incorrect. But aside from the results, your perception of them are subjective even if they are correct. I won't even disagree that ZvT is not ZerO's best match up. ZerO was known more for his ZvP. For both players, their games with Fantasy makes up a large portion of their ZvT record (14 games for ZerO and 7 games for Soulkey). For Soulkey in particular, his dominance against Fantasy is one of the first things that are brought up when people discuss his predator-prey relationships with other players. ZerO vs Terran in standard leagues: 57W 51L (52.78%) ZerO vs Non-Fantasy Terran in standard leagues: 52W 42L (55.3%) Soulkey vs Terran in standard leagues: 36W 31L (52.73%) Soulkey vs Non-Fantasy Terran in standard leagues: 30W 30L (50%) As far as the Korean comments, I am Korean so I am just speaking about discussions with other Koreans plus my perception of how each player was pumped up to the viewer by the Korean commentators (rather than a direct comparison). Finally, speaking as a professional mathematician, I think that it is strange for 1) these number of games to be considered a large sample for statistical evaluation and 2) for a 5~7% difference in winning percentages to be considered meaningful over 100 games. However, this is also why I am willing to accept that I could be wrong and Soulkey could have been just unlucky in individual leagues. EDIT: One area of disagreement: I don't think it is fair to describe ZerO's results in individual leagues as merely decent. PS: I will say that I am a little biased because ZerO was my favorite Zerg at the end of BW. Even though Jaedong was obviously the best Zerg by far, I just liked ZerO more for reasons I can't really explain.
Yes, I did actually did previously consider perception as of primary importance, in addition to accurate memory, and likely should have mentioned it as well. And yes judgement of quality would be subjective to some extent, even if nevertheless accurate. Specifically the item of accurate "objective" perception (probably of both quantitative/qualitative factors) is likely the primary factor at this point, based on the progression of the discussion. (We can set aside quality for the most part, I will however, add one relevant point: some particular aspects regarding game quality, while not being especially concrete, are not necessarily necessarily completely subjective either: For example, does the player control the game from start to finish (e.g. better than other players would be able to in the same situations), or was there the potential or actual occurence of counterplay from the opponent. How fast/accurate were certain executions in situations where the optimal response is known, etc. There may be better examples actually. But anyways, in this type of assessments, over a number of games, there are a considerable number of data (per game), and I would say, in many of these such aspects I noticed Soulkey had a distinct edge. Obviously perception will still apply to some extent - specific significance questionable.)
Regarding the sample size: yes, from a general statistical standpoint, the number of games in question, is not exceedingly large. But in this specific circumstance (e.g. BW, the time period/players in question) they may retain meaningful value.
1) value of statistics regarding Fantasy: Having a relatively large sample size against Fantasy is not necessarily bad in a sense, and retains some measure of value. 1) Fantasy is one of the best Terrans overall, and has pretty good ZvT results, including good performances against a high level Zerg like Jaedong. Based on this fact, a very good showing against Fantasy - even in a smallish ~7 game sample is indicative of very good ZvT ability (possibly, but unlikely to be completely specific to playing Fantasy) - even in complete isolation. ie. the expectation that any zerg, nevermind one who is not especially skilled at ZvT, would beat Fantasy 6-1 in a 7 game series, is essentially 0. Sure the large sample size vs. one player is not necessarily optimal, but in this specific situation (i.e. who was better player/ZvT) I don't know if is necessarily too damaging.
2) regarding Soulkey's ZvT winrate: I think it would be agreeable to say that Soulkey was a comparatively inexperienced player, who was showing relatively dramatic and continuous improvement over time until 2012, so his overall (ZvT) results are less meaningful than for someone like Zero (who has a lot more experience, and who's ZvT winrate stats have been relatively consistent) - Soulkey overall ZvT winrate is something like 53%, but if you look at his last 33 games (arbitrary sure) Record: 22 wins - 11 losses (66.67%). I think this is really more of a relevant indication of his ability, than Fantasy/non-fantasy distinction.I don't think it's necessarily reasonable to say that the 66% is a statistical outlier (perhaps qualitative input would be required, or perhaps not), or that fantasy/non-fantasy is necessarily very meaningful in comparison. Hence, why I originally thought it would be more to the point to go down different lines of discussion.
3) Regarding the significance of 5-7% winrate differentials: Just to establish some context: The 5% for some reason seems to have significance in BW - e.g. difference between B - A - S class player. So theoretically if there is a objective 5-7%+ differential in skill;- I am assuming it is agreeable to say that this would be indicative of a significant difference in skill (e.g. potentially even B, A, or S, depending on the situation). Anyways, for the same reasons mentioned before e.g. the possibility of a player attaining the winrate statistics that Soulkey achieved, but not having the requisite skills seems fairly low percentagewise. Also consider that while Soulkey is not a particularly an experienced player, he is not particularly inexperienced either, especially given the quality of competition he has faced. If anything, a newer player who does not have the requisite skills to achieve/maintain a very high winrate, is probably going to get crunched pretty badly over a 100 game sample, and certainly you would not expect such a player to dramatically improve winrate with time (obviously a factor would be the magnitude of the difference). I guess the other side of the coin is, how often will it be accurate to say that a player who is 5-7%+ better over a 100 game sample, is actually the worse player,..
So how exactly meaningful/accurate are the 100 game winrate statistics specifically in this context (and perhaps in other 100 game samples)? I'm not sure, but my initial feeling was that they generally will be meaningful.
Obviously the # of tournament games will be a fraction of 100, whatever the implications.
Perhaps if there is no specific comparison, regarding Korean commentary, then this could be put aside. Although I am still interested in knowing what the direct comparison would have yielded, in that time period, if something comes up.
If Zero's Individual League accomplishments were not merely decent, I won't argue that point.
It's interesting you mention that Zero is your favorite player - I admit I found the initial characterization you posed of Zero and Soulkey and Sea, and the subsequent response pattern slightly notable for some reason I could not immediately identify. But that is not necessarily here or there. I don't think I have any significant personal bias in this case. Soulkey was not originally a notable player to me. But after watching him improve so much - especially vs T, I became a fan - for some reason I like players that have high ZvT skill. Look up Yellow[arnc]s record against every TvZ "specialist" you can think of - he has a positive record against essentially every one. I wouldn't say Soulkey was ever became of my top favorite players (he may have eventually but who knows), but essentially by force of play I came to see him as the best Zerg in the world at one point.
|
OMG, Midas is back???
He is fucking cute, rooting for him FOR SURE!
|
when is group B? it was on calendar now its gone
|
Magic Woods9326 Posts
On January 10 2014 14:02 dRaW wrote:
when is group B? it was on calendar now its gone
All live Korean casts are on Thursdays and Saturdays
|
+ Show Spoiler +On January 10 2014 10:40 miercat wrote:Show nested quote +On January 10 2014 08:24 lemmata wrote:On January 10 2014 07:37 miercat wrote:+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW. Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on. Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%). I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question. Let's set aside the quality of play for a second, because that can be subjective. I don't think you have any reason to be disturbed. I do not think that your excellent memory is incorrect. But aside from the results, your perception of them are subjective even if they are correct. I won't even disagree that ZvT is not ZerO's best match up. ZerO was known more for his ZvP. For both players, their games with Fantasy makes up a large portion of their ZvT record (14 games for ZerO and 7 games for Soulkey). For Soulkey in particular, his dominance against Fantasy is one of the first things that are brought up when people discuss his predator-prey relationships with other players. ZerO vs Terran in standard leagues: 57W 51L (52.78%) ZerO vs Non-Fantasy Terran in standard leagues: 52W 42L (55.3%) Soulkey vs Terran in standard leagues: 36W 31L (52.73%) Soulkey vs Non-Fantasy Terran in standard leagues: 30W 30L (50%) As far as the Korean comments, I am Korean so I am just speaking about discussions with other Koreans plus my perception of how each player was pumped up to the viewer by the Korean commentators (rather than a direct comparison). Finally, speaking as a professional mathematician, I think that it is strange for 1) these number of games to be considered a large sample for statistical evaluation and 2) for a 5~7% difference in winning percentages to be considered meaningful over 100 games. However, this is also why I am willing to accept that I could be wrong and Soulkey could have been just unlucky in individual leagues. EDIT: One area of disagreement: I don't think it is fair to describe ZerO's results in individual leagues as merely decent. PS: I will say that I am a little biased because ZerO was my favorite Zerg at the end of BW. Even though Jaedong was obviously the best Zerg by far, I just liked ZerO more for reasons I can't really explain. Yes, I did actually did previously consider perception as of primary importance, in addition to accurate memory, and likely should have mentioned it as well. And yes judgement of quality would be subjective to some extent, even if nevertheless accurate. Specifically the item of accurate "objective" perception (probably of both quantitative/qualitative factors) is likely the primary factor at this point, based on the progression of the discussion. (We can set aside quality for the most part, I will however, add one relevant point: some particular aspects regarding game quality, while not being especially concrete, are not necessarily necessarily completely subjective either: For example, does the player control the game from start to finish (e.g. better than other players would be able to in the same situations), or was there the potential or actual occurence of counterplay from the opponent. How fast/accurate were certain executions in situations where the optimal response is known, etc. There may be better examples actually. But anyways, in this type of assessments, over a number of games, there are a considerable number of data (per game), and I would say, in many of these such aspects I noticed Soulkey had a distinct edge. Obviously perception will still apply to some extent - specific significance questionable.) Regarding the sample size: yes, from a general statistical standpoint, the number of games in question, is not exceedingly large. But in this specific circumstance (e.g. BW, the time period/players in question) they may retain meaningful value. 1) value of statistics regarding Fantasy: Having a relatively large sample size against Fantasy is not necessarily bad in a sense, and retains some measure of value. 1) Fantasy is one of the best Terrans overall, and has pretty good ZvT results, including good performances against a high level Zerg like Jaedong. Based on this fact, a very good showing against Fantasy - even in a smallish ~7 game sample is indicative of very good ZvT ability (possibly, but unlikely to be completely specific to playing Fantasy) - even in complete isolation. ie. the expectation that any zerg, nevermind one who is not especially skilled at ZvT, would beat Fantasy 6-1 in a 7 game series, is essentially 0. Sure the large sample size vs. one player is not necessarily optimal, but in this specific situation (i.e. who was better player/ZvT) I don't know if is necessarily too damaging. 2) regarding Soulkey's ZvT winrate: I think it would be agreeable to say that Soulkey was a comparatively inexperienced player, who was showing relatively dramatic and continuous improvement over time until 2012, so his overall (ZvT) results are less meaningful than for someone like Zero (who has a lot more experience, and who's ZvT winrate stats have been relatively consistent) - Soulkey overall ZvT winrate is something like 53%, but if you look at his last 33 games (arbitrary sure) Record: 22 wins - 11 losses (66.67%). I think this is really more of a relevant indication of his ability, than Fantasy/non-fantasy distinction.I don't think it's necessarily reasonable to say that the 66% is a statistical outlier (perhaps qualitative input would be required, or perhaps not), or that fantasy/non-fantasy is necessarily very meaningful in comparison. Hence, why I originally thought it would be more to the point to go down different lines of discussion. 3) Regarding the significance of 5-7% winrate differentials: Just to establish some context: The 5% for some reason seems to have significance in BW - e.g. difference between B - A - S class player. So theoretically if there is a objective 5-7%+ differential in skill;- I am assuming it is agreeable to say that this would be indicative of a significant difference in skill (e.g. potentially even B, A, or S, depending on the situation). Anyways, for the same reasons mentioned before e.g. the possibility of a player attaining the winrate statistics that Soulkey achieved, but not having the requisite skills seems fairly low percentagewise. Also consider that while Soulkey is not a particularly an experienced player, he is not particularly inexperienced either, especially given the quality of competition he has faced. If anything, a newer player who does not have the requisite skills to achieve/maintain a very high winrate, is probably going to get crunched pretty badly over a 100 game sample, and certainly you would not expect such a player to dramatically improve winrate with time (obviously a factor would be the magnitude of the difference). I guess the other side of the coin is, how often will it be accurate to say that a player who is 5-7%+ better over a 100 game sample, is actually the worse player,.. So how exactly meaningful/accurate are the 100 game winrate statistics specifically in this context (and perhaps in other 100 game samples)? I'm not sure, but my initial feeling was that they generally will be meaningful. Obviously the # of tournament games will be a fraction of 100, whatever the implications. Perhaps if there is no specific comparison, regarding Korean commentary, then this could be put aside. Although I am still interested in knowing what the direct comparison would have yielded, in that time period, if something comes up. If Zero's Individual League accomplishments were not merely decent, I won't argue that point. It's interesting you mention that Zero is your favorite player - I admit I found the initial characterization you posed of Zero and Soulkey and Sea, and the subsequent response pattern slightly notable for some reason I could not immediately identify. But that is not necessarily here or there. I don't think I have any significant personal bias in this case. Soulkey was not originally a notable player to me. But after watching him improve so much - especially vs T, I became a fan - for some reason I like players that have high ZvT skill. Look up Yellow[arnc]s record against every TvZ "specialist" you can think of - he has a positive record against essentially every one. I wouldn't say Soulkey was ever became of my top favorite players (he may have eventually but who knows), but essentially by force of play I came to see him as the best Zerg in the world at one point.
I will start with the conclusion: There's no way you can actually prove the claim that Soulkey was a class above ZerO. There's no way that I can prove the opposite. I don't think you disagree based on what I've read---unless you think that 33 games can prove someone's skill in objective terms. What follows is just bar talk.
You can make the case that the majority perception and their criteria are wrong. That may all be true. However, you shouldn't be surprised that ZerO was considered an S-class player because most people go by OSL/MSL performance when they rate players. In terms of OSL/MSL performance, Soulkey barely has a leg to stand on, while ZerO has a long list of accomplishments beyond anything Soulkey achieved in those leagues. Again, that's unfair, but you can't be surprised about the perception itself even if you disagree with its conclusion.
The reason I chose to go by individual tournament records is because that is what the players themselves care about the most. When SKT made Bisu focus on ProLeague practice he even took the unprecedented step of complaining publicly that it was hurting his individual league performance. I've already indicated that considering individual league performance is not somehow more scientific. It's just that I've chosen to focus on the games that mattered the most. That doesn't make it right, but this is what people do when the numbers don't yield a clear answer. There are also BOx type matches in individual leagues, which requires an interesting extra layer of planning.
About Korean commentary, the most specific quote I can remember is the one I included in my earlier post. Kim Carry said that "ZerO is a player who never won a championship, but was worthy of one." That's pretty much the highest compliment you can give someone who did not win either league. My perception comes from statements of that kind often being made about ZerO but never about Soulkey. I noticed your comment about ZerO never being S-class, but he was actually the one being referred to as S-class on TV. Most of this is really just due to the aforementioned Korean bias in favor of individual leagues---which might be unfair, but it is what it is. Sea had a higher peak ELO than both ZerO and Soulkey, but he was never considered S-class precisely because of his individual league performance. Even now Kim Carry will point out that Sea was always missing that extra 2% needed to become a champion.
|
On January 10 2014 14:39 lemmata wrote:+ Show Spoiler +On January 10 2014 10:40 miercat wrote:Show nested quote +On January 10 2014 08:24 lemmata wrote:On January 10 2014 07:37 miercat wrote:+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW. Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on. Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%). I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question. Let's set aside the quality of play for a second, because that can be subjective. I don't think you have any reason to be disturbed. I do not think that your excellent memory is incorrect. But aside from the results, your perception of them are subjective even if they are correct. I won't even disagree that ZvT is not ZerO's best match up. ZerO was known more for his ZvP. For both players, their games with Fantasy makes up a large portion of their ZvT record (14 games for ZerO and 7 games for Soulkey). For Soulkey in particular, his dominance against Fantasy is one of the first things that are brought up when people discuss his predator-prey relationships with other players. ZerO vs Terran in standard leagues: 57W 51L (52.78%) ZerO vs Non-Fantasy Terran in standard leagues: 52W 42L (55.3%) Soulkey vs Terran in standard leagues: 36W 31L (52.73%) Soulkey vs Non-Fantasy Terran in standard leagues: 30W 30L (50%) As far as the Korean comments, I am Korean so I am just speaking about discussions with other Koreans plus my perception of how each player was pumped up to the viewer by the Korean commentators (rather than a direct comparison). Finally, speaking as a professional mathematician, I think that it is strange for 1) these number of games to be considered a large sample for statistical evaluation and 2) for a 5~7% difference in winning percentages to be considered meaningful over 100 games. However, this is also why I am willing to accept that I could be wrong and Soulkey could have been just unlucky in individual leagues. EDIT: One area of disagreement: I don't think it is fair to describe ZerO's results in individual leagues as merely decent. PS: I will say that I am a little biased because ZerO was my favorite Zerg at the end of BW. Even though Jaedong was obviously the best Zerg by far, I just liked ZerO more for reasons I can't really explain. Yes, I did actually did previously consider perception as of primary importance, in addition to accurate memory, and likely should have mentioned it as well. And yes judgement of quality would be subjective to some extent, even if nevertheless accurate. Specifically the item of accurate "objective" perception (probably of both quantitative/qualitative factors) is likely the primary factor at this point, based on the progression of the discussion. (We can set aside quality for the most part, I will however, add one relevant point: some particular aspects regarding game quality, while not being especially concrete, are not necessarily completely subjective either: For example, does the player control the game from start to finish (e.g. better than other players would be able to in the same situations), or was there the potential or actual occurence of counterplay from the opponent. How fast/accurate were certain executions in situations where the optimal response is known, etc. There may be better examples actually. But anyways, in this type of assessments, over a number of games, there are a considerable number of data (per game), and I would say, in many of these such aspects I noticed Soulkey had a distinct edge. Obviously perception will still apply to some extent - specific significance questionable.) Regarding the sample size: yes, from a general statistical standpoint, the number of games in question, is not exceedingly large. But in this specific circumstance (e.g. BW, the time period/players in question) they may retain meaningful value. 1) value of statistics regarding Fantasy: Having a relatively large sample size against Fantasy is not necessarily bad in a sense, and retains some measure of value. 1) Fantasy is one of the best Terrans overall, and has pretty good ZvT results, including good performances against a high level Zerg like Jaedong. Based on this fact, a very good showing against Fantasy - even in a smallish ~7 game sample is indicative of very good ZvT ability (possibly, but unlikely to be completely specific to playing Fantasy) - even in complete isolation. ie. the expectation that any zerg, nevermind one who is not especially skilled at ZvT, would beat Fantasy 6-1 in a 7 game series, is essentially 0. Sure the large sample size vs. one player is not necessarily optimal, but in this specific situation (i.e. who was better player/ZvT) I don't know if is necessarily too damaging. 2) regarding Soulkey's ZvT winrate: I think it would be agreeable to say that Soulkey was a comparatively inexperienced player, who was showing relatively dramatic and continuous improvement over time until 2012, so his overall (ZvT) results are less meaningful than for someone like Zero (who has a lot more experience, and who's ZvT winrate stats have been relatively consistent) - Soulkey overall ZvT winrate is something like 53%, but if you look at his last 33 games (arbitrary sure) Record: 22 wins - 11 losses (66.67%). I think this is really more of a relevant indication of his ability, than Fantasy/non-fantasy distinction.I don't think it's necessarily reasonable to say that the 66% is a statistical outlier (perhaps qualitative input would be required, or perhaps not), or that fantasy/non-fantasy is necessarily very meaningful in comparison. Hence, why I originally thought it would be more to the point to go down different lines of discussion. 3) Regarding the significance of 5-7% winrate differentials: Just to establish some context: The 5% for some reason seems to have significance in BW - e.g. difference between B - A - S class player. So theoretically if there is a objective 5-7%+ differential in skill;- I am assuming it is agreeable to say that this would be indicative of a significant difference in skill (e.g. potentially even B, A, or S, depending on the situation). Anyways, for the same reasons mentioned before e.g. the possibility of a player attaining the winrate statistics that Soulkey achieved, but not having the requisite skills seems fairly low percentagewise. Also consider that while Soulkey is not a particularly an experienced player, he is not particularly inexperienced either, especially given the quality of competition he has faced. If anything, a newer player who does not have the requisite skills to achieve/maintain a very high winrate, is probably going to get crunched pretty badly over a 100 game sample, and certainly you would not expect such a player to dramatically improve winrate with time (obviously a factor would be the magnitude of the difference). I guess the other side of the coin is, how often will it be accurate to say that a player who is 5-7%+ better over a 100 game sample, is actually the worse player,.. So how exactly meaningful/accurate are the 100 game winrate statistics specifically in this context (and perhaps in other 100 game samples)? I'm not sure, but my initial feeling was that they generally will be meaningful. Obviously the # of tournament games will be a fraction of 100, whatever the implications. Perhaps if there is no specific comparison, regarding Korean commentary, then this could be put aside. Although I am still interested in knowing what the direct comparison would have yielded, in that time period, if something comes up. If Zero's Individual League accomplishments were not merely decent, I won't argue that point. It's interesting you mention that Zero is your favorite player - I admit I found the initial characterization you posed of Zero and Soulkey and Sea, and the subsequent response pattern slightly notable for some reason I could not immediately identify. But that is not necessarily here or there. I don't think I have any significant personal bias in this case. Soulkey was not originally a notable player to me. But after watching him improve so much - especially vs T, I became a fan - for some reason I like players that have high ZvT skill. Look up Yellow[arnc]s record against every TvZ "specialist" you can think of - he has a positive record against essentially every one. I wouldn't say Soulkey was ever became of my top favorite players (he may have eventually but who knows), but essentially by force of play I came to see him as the best Zerg in the world at one point. I will start with the conclusion: There's no way you can actually prove the claim that Soulkey was a class above ZerO. There's no way that I can prove the opposite. I don't think you disagree based on what I've read---unless you think that 33 games can prove someone's skill in objective terms. What follows is just bar talk. You can make the case that the majority perception and their criteria are wrong. That may all be true. However, you shouldn't be surprised that ZerO was considered an S-class player because most people go by OSL/MSL performance when they rate players. In terms of OSL/MSL performance, Soulkey barely has a leg to stand on, while ZerO has a long list of accomplishments beyond anything Soulkey achieved in those leagues. Again, that's unfair, but you can't be surprised about the perception itself even if you disagree with its conclusion. The reason I chose to go by individual tournament records is because that is what the players themselves care about the most. When SKT made Bisu focus on ProLeague practice he even took the unprecedented step of complaining publicly that it was hurting his individual league performance. I've already indicated that considering individual league performance is not somehow more scientific. It's just that I've chosen to focus on the games that mattered the most. That doesn't make it right, but this is what people do when the numbers don't yield a clear answer. There are also BOx type matches in individual leagues, which requires an interesting extra layer of planning. About Korean commentary, the most specific quote I can remember is the one I included in my earlier post. Kim Carry said that "ZerO is a player who never won a championship, but was worthy of one." That's pretty much the highest compliment you can give someone who did not win either league. My perception comes from statements of that kind often being made about ZerO but never about Soulkey. I noticed your comment about ZerO never being S-class, but he was actually the one being referred to as S-class on TV. Most of this is really just due to the aforementioned Korean bias in favor of individual leagues---which might be unfair, but it is what it is. Sea had a higher peak ELO than both ZerO and Soulkey, but he was never considered S-class precisely because of his individual league performance. Even now Kim Carry will point out that Sea was always missing that extra 2% needed to become a champion.
That specific conclusion is not particularly relevant. To clarify: I wasn't necessarily claiming Soulkey was class above Zero, but rather that even a seemingly small difference (assuming it's valid) of 5-7%, would actually be quite significant in terms of relative skill difference - whether it is specifically a class above, was not particularly important. (I did mention that the statement regarding class- was just to establish context)
Regarding the sample size, in certain cases, a smaller and/or more recent, sample size can actually be more meaningful, than larger more lengthy ones. Perhaps the relevant one is Soulkey's mediocre ZvT overall record ~53%, in comparison with the most recent run of 22-11 66%. No one can reasonably claim that Soulkey in the last couple years was playing at a 53% ZvT skill level. Sure if he was somehow luckboxing/cheesing/playing bad players, etc. maybe the validity of his skill could be questioned, but given the fact that he was a relatively young, constantly improving player, the recent run of 66%, it's pretty clear that considering the larger, more lengthy sample size was actually less meaningful - in fact, highly inaccurate for the discussion at hand.
Perhaps it is true that players care more about Individual League. But think about the context. These players are practicing 12 hours a day, often with Proleague success prioritized over Individual League, and they get to play on televised games very rarely compared to the amount of practice they are putting in. They are going to play at a very high level (or try to) whenever it is an official match. It is not wrong to put some degree more importance on Individual results, but I don't think that in light of all the facts, that Individual results retain so much importance in this case: If anything the most accurate way of determining player skill - is the series of inhouse ranking tournies, etc. This is often the primary measure of skill with regards to those directly involved in the Pro Scene. Consider cases, like Jaehoon - most people think he is a terrible player, but the fact is he regularly did exceptionally well in inhouse ranking tournies - indicating very high skill - which is why the coach often put him in ace match situations, even though his results were very poor officially. Additionally, many players highly respect Jaehoon's ability (mentioned in interviews). This is not even an issue with Soulkey - He is consistently winning more inouse tournies than Zero, Zero thinks Soulkey is better, Soulkey has much better statistics in every matchup ----buut Zero better Invididual League results - It's not really a hard computation - I am not discriminating against "popular opinion," more so the goal is just to get an objective view of who the best player is. Again, even if you want to perhaps correctly prioritize Individual League games, and assign more weight to them - it is still not nearly enough to discount all the other significant factors in play here (regarding who is the better player).
I don't speak Korean, but this is the first time I have heard anyone imply that Zero was an S-Class Player. Obviously I don't know the specific TV quote you were referring to: maybe it was just an incidental discussion of S-class and Zero, without a direct assertion that Zero was in fact S-Class (which is completely different).
But truly, that would be absolutely ridiculous. I fear the discussion is starting to devolve significantly at this point. Let me get to the point. Even if you want to arbitrarily focus on Starleague results, here is what S-Class level results look like:
Jaedong:
Winner: 2010 WCG Korea Finals 10 games
Runner-Up: 2010 Korean Air OnGameNet Starleague Season 2 18 games
Runner-Up: 2010 Bigfile MBCGame StarCraft League 19 games
Runner-Up: 2010 Hana Daetoo Securities MBCGame StarCraft League 16 games
Winner: 2009-2010 NATE MBCGame StarCraft League 14 games
Third: 2009 WCG Korea Finals 10 games
Winner: 2009 Bacchus OnGameNet Starleague 12 games
Winner: 2009 Averatec-Intel Classic Special Match 5 games
Winner: 2008-2009 Batoo OnGameNet Starleague 19 games
Winner: 2008 WCG Korea Finals 9 games
Winner: 2008 GOMTV TG-Intel Classic Season 1 17 games
Runner-Up: 2008 Arena MBCGame StarCraft League 15 games
Winner: 2008 GOMTV MBCGame StarCraft League Season 4 16 games
Winner: 2007 EVER OnGameNet Starleague 13 games
Winner: 2007 Seoul International e-Sports Festival 19 games
Refer to also: Flash, Bisu, Stork, etc.
Here is what S-class Tourny results don't look like:
Zero:
Runner-Up: 2011 ABC Mart MBCGame StarCraft League 16 games
It's such a ridiculous comparison, I don't even wan't to comment on it much further. But suffice to say, I think that anyone could legitimately be surprised that Zero was supposedly considered S-Class(which really I still doubt specifically happened), even if you only want to only look at Starleague results.
Again on the balance of all the information provided (including potentially accurate higher weighting of Starleague games), I think the position that Soulkey was the better player (whether it is specifically a class or more or less is not too important), is much stronger. I think I have provided most of the strongest/relevant arguments already, and I don't really think there are many more relevant/significant points to add (to further the discussion), or that there is the need to do so.
|
Magic Woods9326 Posts
PS. I never considered either even close to S-Class.
|
+ Show Spoiler +On January 11 2014 04:37 miercat wrote:Show nested quote +On January 10 2014 14:39 lemmata wrote:+ Show Spoiler +On January 10 2014 10:40 miercat wrote:Show nested quote +On January 10 2014 08:24 lemmata wrote:On January 10 2014 07:37 miercat wrote:+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW. Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on. Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%). I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question. Let's set aside the quality of play for a second, because that can be subjective. I don't think you have any reason to be disturbed. I do not think that your excellent memory is incorrect. But aside from the results, your perception of them are subjective even if they are correct. I won't even disagree that ZvT is not ZerO's best match up. ZerO was known more for his ZvP. For both players, their games with Fantasy makes up a large portion of their ZvT record (14 games for ZerO and 7 games for Soulkey). For Soulkey in particular, his dominance against Fantasy is one of the first things that are brought up when people discuss his predator-prey relationships with other players. ZerO vs Terran in standard leagues: 57W 51L (52.78%) ZerO vs Non-Fantasy Terran in standard leagues: 52W 42L (55.3%) Soulkey vs Terran in standard leagues: 36W 31L (52.73%) Soulkey vs Non-Fantasy Terran in standard leagues: 30W 30L (50%) As far as the Korean comments, I am Korean so I am just speaking about discussions with other Koreans plus my perception of how each player was pumped up to the viewer by the Korean commentators (rather than a direct comparison). Finally, speaking as a professional mathematician, I think that it is strange for 1) these number of games to be considered a large sample for statistical evaluation and 2) for a 5~7% difference in winning percentages to be considered meaningful over 100 games. However, this is also why I am willing to accept that I could be wrong and Soulkey could have been just unlucky in individual leagues. EDIT: One area of disagreement: I don't think it is fair to describe ZerO's results in individual leagues as merely decent. PS: I will say that I am a little biased because ZerO was my favorite Zerg at the end of BW. Even though Jaedong was obviously the best Zerg by far, I just liked ZerO more for reasons I can't really explain. Yes, I did actually did previously consider perception as of primary importance, in addition to accurate memory, and likely should have mentioned it as well. And yes judgement of quality would be subjective to some extent, even if nevertheless accurate. Specifically the item of accurate "objective" perception (probably of both quantitative/qualitative factors) is likely the primary factor at this point, based on the progression of the discussion. (We can set aside quality for the most part, I will however, add one relevant point: some particular aspects regarding game quality, while not being especially concrete, are not necessarily completely subjective either: For example, does the player control the game from start to finish (e.g. better than other players would be able to in the same situations), or was there the potential or actual occurence of counterplay from the opponent. How fast/accurate were certain executions in situations where the optimal response is known, etc. There may be better examples actually. But anyways, in this type of assessments, over a number of games, there are a considerable number of data (per game), and I would say, in many of these such aspects I noticed Soulkey had a distinct edge. Obviously perception will still apply to some extent - specific significance questionable.) Regarding the sample size: yes, from a general statistical standpoint, the number of games in question, is not exceedingly large. But in this specific circumstance (e.g. BW, the time period/players in question) they may retain meaningful value. 1) value of statistics regarding Fantasy: Having a relatively large sample size against Fantasy is not necessarily bad in a sense, and retains some measure of value. 1) Fantasy is one of the best Terrans overall, and has pretty good ZvT results, including good performances against a high level Zerg like Jaedong. Based on this fact, a very good showing against Fantasy - even in a smallish ~7 game sample is indicative of very good ZvT ability (possibly, but unlikely to be completely specific to playing Fantasy) - even in complete isolation. ie. the expectation that any zerg, nevermind one who is not especially skilled at ZvT, would beat Fantasy 6-1 in a 7 game series, is essentially 0. Sure the large sample size vs. one player is not necessarily optimal, but in this specific situation (i.e. who was better player/ZvT) I don't know if is necessarily too damaging. 2) regarding Soulkey's ZvT winrate: I think it would be agreeable to say that Soulkey was a comparatively inexperienced player, who was showing relatively dramatic and continuous improvement over time until 2012, so his overall (ZvT) results are less meaningful than for someone like Zero (who has a lot more experience, and who's ZvT winrate stats have been relatively consistent) - Soulkey overall ZvT winrate is something like 53%, but if you look at his last 33 games (arbitrary sure) Record: 22 wins - 11 losses (66.67%). I think this is really more of a relevant indication of his ability, than Fantasy/non-fantasy distinction.I don't think it's necessarily reasonable to say that the 66% is a statistical outlier (perhaps qualitative input would be required, or perhaps not), or that fantasy/non-fantasy is necessarily very meaningful in comparison. Hence, why I originally thought it would be more to the point to go down different lines of discussion. 3) Regarding the significance of 5-7% winrate differentials: Just to establish some context: The 5% for some reason seems to have significance in BW - e.g. difference between B - A - S class player. So theoretically if there is a objective 5-7%+ differential in skill;- I am assuming it is agreeable to say that this would be indicative of a significant difference in skill (e.g. potentially even B, A, or S, depending on the situation). Anyways, for the same reasons mentioned before e.g. the possibility of a player attaining the winrate statistics that Soulkey achieved, but not having the requisite skills seems fairly low percentagewise. Also consider that while Soulkey is not a particularly an experienced player, he is not particularly inexperienced either, especially given the quality of competition he has faced. If anything, a newer player who does not have the requisite skills to achieve/maintain a very high winrate, is probably going to get crunched pretty badly over a 100 game sample, and certainly you would not expect such a player to dramatically improve winrate with time (obviously a factor would be the magnitude of the difference). I guess the other side of the coin is, how often will it be accurate to say that a player who is 5-7%+ better over a 100 game sample, is actually the worse player,.. So how exactly meaningful/accurate are the 100 game winrate statistics specifically in this context (and perhaps in other 100 game samples)? I'm not sure, but my initial feeling was that they generally will be meaningful. Obviously the # of tournament games will be a fraction of 100, whatever the implications. Perhaps if there is no specific comparison, regarding Korean commentary, then this could be put aside. Although I am still interested in knowing what the direct comparison would have yielded, in that time period, if something comes up. If Zero's Individual League accomplishments were not merely decent, I won't argue that point. It's interesting you mention that Zero is your favorite player - I admit I found the initial characterization you posed of Zero and Soulkey and Sea, and the subsequent response pattern slightly notable for some reason I could not immediately identify. But that is not necessarily here or there. I don't think I have any significant personal bias in this case. Soulkey was not originally a notable player to me. But after watching him improve so much - especially vs T, I became a fan - for some reason I like players that have high ZvT skill. Look up Yellow[arnc]s record against every TvZ "specialist" you can think of - he has a positive record against essentially every one. I wouldn't say Soulkey was ever became of my top favorite players (he may have eventually but who knows), but essentially by force of play I came to see him as the best Zerg in the world at one point. I will start with the conclusion: There's no way you can actually prove the claim that Soulkey was a class above ZerO. There's no way that I can prove the opposite. I don't think you disagree based on what I've read---unless you think that 33 games can prove someone's skill in objective terms. What follows is just bar talk. You can make the case that the majority perception and their criteria are wrong. That may all be true. However, you shouldn't be surprised that ZerO was considered an S-class player because most people go by OSL/MSL performance when they rate players. In terms of OSL/MSL performance, Soulkey barely has a leg to stand on, while ZerO has a long list of accomplishments beyond anything Soulkey achieved in those leagues. Again, that's unfair, but you can't be surprised about the perception itself even if you disagree with its conclusion. The reason I chose to go by individual tournament records is because that is what the players themselves care about the most. When SKT made Bisu focus on ProLeague practice he even took the unprecedented step of complaining publicly that it was hurting his individual league performance. I've already indicated that considering individual league performance is not somehow more scientific. It's just that I've chosen to focus on the games that mattered the most. That doesn't make it right, but this is what people do when the numbers don't yield a clear answer. There are also BOx type matches in individual leagues, which requires an interesting extra layer of planning. About Korean commentary, the most specific quote I can remember is the one I included in my earlier post. Kim Carry said that "ZerO is a player who never won a championship, but was worthy of one." That's pretty much the highest compliment you can give someone who did not win either league. My perception comes from statements of that kind often being made about ZerO but never about Soulkey. I noticed your comment about ZerO never being S-class, but he was actually the one being referred to as S-class on TV. Most of this is really just due to the aforementioned Korean bias in favor of individual leagues---which might be unfair, but it is what it is. Sea had a higher peak ELO than both ZerO and Soulkey, but he was never considered S-class precisely because of his individual league performance. Even now Kim Carry will point out that Sea was always missing that extra 2% needed to become a champion. That specific conclusion is not particularly relevant. To clarify: I wasn't necessarily claiming Soulkey was class above Zero, but rather that even a seemingly small difference (assuming it's valid) of 5-7%, would actually be quite significant in terms of relative skill difference - whether it is specifically a class above, was not particularly important. (I did mention that the statement regarding class- was just to establish context) Regarding the sample size, in certain cases, a smaller and/or more recent, sample size can actually be more meaningful, than larger more lengthy ones. Perhaps the relevant one is Soulkey's mediocre ZvT overall record ~53%, in comparison with the most recent run of 22-11 66%. No one can reasonably claim that Soulkey in the last couple years was playing at a 53% ZvT skill level. Sure if he was somehow luckboxing/cheesing/playing bad players, etc. maybe the validity of his skill could be questioned, but given the fact that he was a relatively young, constantly improving player, the recent run of 66%, it's pretty clear that considering the larger, more lengthy sample size was actually less meaningful - in fact, highly inaccurate for the discussion at hand. Perhaps it is true that players care more about Individual League. But think about the context. These players are practicing 12 hours a day, often with Proleague success prioritized over Individual League, and they get to play on televised games very rarely compared to the amount of practice they are putting in. They are going to play at a very high level (or try to) whenever it is an official match. It is not wrong to put some degree more importance on Individual results, but I don't think that in light of all the facts, that Individual results retain so much importance in this case: If anything the most accurate way of determining player skill - is the series of inhouse ranking tournies, etc. This is often the primary measure of skill with regards to those directly involved in the Pro Scene. Consider cases, like Jaehoon - most people think he is a terrible player, but the fact is he regularly did exceptionally well in inhouse ranking tournies - indicating very high skill - which is why the coach often put him in ace match situations, even though his results were very poor officially. Additionally, many players highly respect Jaehoon's ability (mentioned in interviews). This is not even an issue with Soulkey - He is consistently winning more inouse tournies than Zero, Zero thinks Soulkey is better, Soulkey has much better statistics in every matchup ----buut Zero better Invididual League results - It's not really a hard computation - I am not discriminating against "popular opinion," more so the goal is just to get an objective view of who the best player is. Again, even if you want to perhaps correctly prioritize Individual League games, and assign more weight to them - it is still not nearly enough to discount all the other significant factors in play here (regarding who is the better player). I don't speak Korean, but this is the first time I have heard anyone imply that Zero was an S-Class Player. Obviously I don't know the specific TV quote you were referring to: maybe it was just an incidental discussion of S-class and Zero, without a direct assertion that Zero was in fact S-Class (which is completely different). But truly, that would be absolutely ridiculous. I fear the discussion is starting to devolve significantly at this point. Let me get to the point. Even if you want to arbitrarily focus on Starleague results, here is what S-Class level results look like: Jaedong: Winner: 2010 WCG Korea Finals 10 games
Runner-Up: 2010 Korean Air OnGameNet Starleague Season 2 18 games
Runner-Up: 2010 Bigfile MBCGame StarCraft League 19 games
Runner-Up: 2010 Hana Daetoo Securities MBCGame StarCraft League 16 games
Winner: 2009-2010 NATE MBCGame StarCraft League 14 games
Third: 2009 WCG Korea Finals 10 games
Winner: 2009 Bacchus OnGameNet Starleague 12 games
Winner: 2009 Averatec-Intel Classic Special Match 5 games
Winner: 2008-2009 Batoo OnGameNet Starleague 19 games
Winner: 2008 WCG Korea Finals 9 games
Winner: 2008 GOMTV TG-Intel Classic Season 1 17 games
Runner-Up: 2008 Arena MBCGame StarCraft League 15 games
Winner: 2008 GOMTV MBCGame StarCraft League Season 4 16 games
Winner: 2007 EVER OnGameNet Starleague 13 games
Winner: 2007 Seoul International e-Sports Festival 19 games
Refer to also: Flash, Bisu, Stork, etc. Here is what S-class Tourny results don't look like: Zero: Runner-Up: 2011 ABC Mart MBCGame StarCraft League 16 gamesIt's such a ridiculous comparison, I don't even wan't to comment on it much further. But suffice to say, I think that anyone could legitimately be surprised that Zero was supposedly considered S-Class(which really I still doubt specifically happened), even if you only want to only look at Starleague results. Again on the balance of all the information provided (including potentially accurate higher weighting of Starleague games), I think the position that Soulkey was the better player (whether it is specifically a class or more or less is not too important), is much stronger. I think I have provided most of the strongest/relevant arguments already, and I don't really think there are many more relevant/significant points to add (to further the discussion), or that there is the need to do so.
LOL. Maybe the difference comes down to the fact that I consider Jaedong God-class, which is beyond S-class. :-) Definitions, definitions... Jaedong is the greatest Zerg of all time...kind of an unfair comparison. Also, let's not boil down Zero's resume to that one line from Liquipedia when I've already given a more complete list.
I am not saying that you can't argue for Soulkey. I don't disagree that you're making a case. It's just not an objective logical proof of anything, although you are somehow beginning to convince me of the conclusion regardless. PM me if you want to talk some more. I don't want to keep bumping the thread with what's turning into a two-person discussion. Having this kind of conversation again really brings back memories. Who do I need to thank for this? Sonic or FIX?
EDIT: long quote spoilered
|
+ Show Spoiler +On January 11 2014 06:48 lemmata wrote:+ Show Spoiler +On January 11 2014 04:37 miercat wrote:Show nested quote +On January 10 2014 14:39 lemmata wrote:+ Show Spoiler +On January 10 2014 10:40 miercat wrote:Show nested quote +On January 10 2014 08:24 lemmata wrote:On January 10 2014 07:37 miercat wrote:+ Show Spoiler +On January 10 2014 07:01 lemmata wrote:Show nested quote +On January 10 2014 05:46 miercat wrote:On January 10 2014 04:44 lemmata wrote:On January 09 2014 15:29 miercat wrote:On January 09 2014 14:34 Epoxide wrote:On January 09 2014 13:46 miercat wrote:
Yes, for some reason I found it very interesting watching the Juice micro during the Group Ceremony. There were at least 6 different movements applied to the various juices (not drinking, just moving), by Sonic and LSW primarily, throughout the night. I never actually saw them drink any juice..
On a different topic, I don't recall Zero being considered at all - nevermind "widely considered" the best Zerg after Jaedong, in the most recent Proleague years. If anything Jaedong was slumping pretty hard, and it was actually Soulkey who was essentially destroying everybody (even Terrans) except Flash, and was considered pretty clearly the best Zerg in the world. Zero never reached the same level that Soulkey did. Do you not think that ZerO's second place finish in ABC Mart, and consistent semifinals appearances in tving and PDPop trumps Souley's round of 16 appearances? Perhaps Soulkey was considered better in ZvT, but I do think that ZerO should be called the better overall player. Individual League performance may be in Zero's favor. But unless I am mistaken, I remember a period in time when Soulkey was considered a favorite against every player in the world except Flash. On the other hand, Zero would have, at all points in time, been considered an underdog to quite a number of players (including many Terran players). Based on overall statistics, and the players Soulkey was defeating, and the style in which he was doing so, it seems like he reached the higher "peak" of performance. And probably at one point he was playing the second highest level of ZvT ever (after Jaedong). Soulkey has a much better overall winrate, a much better ZvP winrate, and a much better ZvZ, winrate, and a better, although misleadingly low overall ZvT winrate (given that he improved his ZvT winrate SO dramatically in the last couple years).Soulkey had significantly less experience than Zero, and yet already significantly better statistics in all matchups. Nevermind just ZvT, I think there is compelling evidence that Soulkey was significantly better in every single matchup - by statistics, but also by directly observing quality of games. At one point, Soulkey was on the verge of being S-Class (or potentially even S-Class), Zero was having some decent success in individual leagues, but his overall play ranged from mediocre - to very good, but never really near S-Class. I would be interested however, in knowing what the Professional Korean commentators had to say about the topic. Soulkey: All: 94-66 (58.75%) vT: 36-31 (53.73%) vZ: 30-20 (60.00%) vP: 28-15 (65.12%) Zero: All: 224-184 (54.90%) vT: 57-51 (52.78%) vZ: 99-87 (53.23%) vP: 68-46 (59.65%) I think that these numbers tell a partial story. As you say, ZerO's win rate is better in individual leagues. How one rates these two players might be a matter of how much weight is given to ProLeague performance. I think the common (but not universal) Korean view is that OSL > MSL > ProLeague. ProLeague performance can sometimes depend a lot on how the player is used and whether targeted sniping occurs and other randomness. Under that view, the following list would tell a different story. ZerO was consistently within sniffing distance of winning individual titles. He was having a lot more than decent success. Soulkey Round of 36(1): Bacchus OSL Round of 16(3): PD POP MSL, ABC Mart MSL, tving OSL ZerO Round of 32(4): GOM TV MSL, Club Day MSL, Hana MSL, Big File MSL Round of 16(1): Nate MSL, Quarterfinal(5): Avalon MSL, Bacchus OSL, EVER OSL, Korean Air OSL 1, Korean Air OSL 2 Semifinal(3): Lost Saga MSL, PD POP MSL, Tving OSL Runner-Up(2): ABC Mart MSL, Blizzcon 2009 Final thought: The people who rank JangBi high on the Protoss hierarchy probably rate ZerO highly as well. That's just a guess. What do you think? I am really excited about JangBi in this tournament. I already conceded that Zero has done better in Individual Leagues, in my previous assessment. Taking that into account, along with all the other factors, I think its still clear that Soulkey was the much better player in every single matchup. The statistical advantage (e.g. 4%, 7%, 6%) is absolutely huge, and along with Soulkey's superior performance against higher level players in general (along with potentially the 2nd highest level of ZvT ever), these facts are just simply too powerful to be discounted by results in Individual Leagues - maybe if the all the relative differences were smaller, but certainly not in this case. And again, on a qualitative level, Zero may have had some "goosebumps" moments, but Soulkey had a consistently much more compelling and impressive style of play, against high level competition, and a higher peak of performance. Zero was not near the same level. Honestly, I was surprised there was any argument at all. It really wasn't even close (based on watching all the relevant Pro BW games, among other things). I am not sure how in depth you were following BW at the time, but maybe you missed many of the games I saw, or other such things... + Show Spoiler + Two people can watch the same game and arrive at different conclusions. I think we'll leave it at that. I do think that my opinion is the far more common one in the Korean community, which doesn't make it right (after all, experts are always smaller in number!) but I think you should at least consider the possibility that your opinion is too strong. Also, the difference between ZerO and Soulkey against top Terrans mostly boils down to the fact that Soulkey owns Fantasy and Fantasy owns ZerO. Meanwhile, I will take your view seriously and consider the possibility that Soulkey had terrible luck in individual leagues. While I compared him to Sea, Soulkey was a level above Sea at the end of BW. Certainly I think there is at least a small possibility that my opinion is too strong. It would be quite disturbing (and probably unlikely) if it was however, based on how accurate my memory usually is for this type of information, and based on the large sample sizes (games, and other info), gathered over a relatively significant timeframe, that I am basing this opinion on. Regarding Fantasy: top level Terrans, perhaps. But more to the point, I think it was evidenced that Soulkey's ZvT was at a much higher level than any zerg except Jaedong(based on the statistics, and by observing quality/style of ZvT wins). Other than Soulkey, no other Zerg was favorite against every Terran except Flash (Soulkey once stated in interview that he was confident against every Terran except Flash - I think he demonstrated it in practical terms as well). And I would actually put a number of Zergs ahead of Zero in ZvT ability. Despite showing a number of notable ZvT games, I don't think he was generally very strong in ZvT e.g. 57-51 (52.78%). I am curious however, specifically what Korean opinions you are referring to. Certainly I wouldn't disagree Zero was at times a very good Zerg player with decent League results - and perhaps deservingly so, but specifically comparing Soulkey's peak performance in terms of skill, to Zero's would be most specifically relevant to the current discussion. I am quite sure that they would have specifically referenced this exact thing at some point, especially considering they were on the same team, and that they were being constantly compared. The fact that Zero thought Soulkey was better is certainly notable, and I have not heard any Korean comments specifically stating that Zero was better than Soulkey, regarding the time period in question. Let's set aside the quality of play for a second, because that can be subjective. I don't think you have any reason to be disturbed. I do not think that your excellent memory is incorrect. But aside from the results, your perception of them are subjective even if they are correct. I won't even disagree that ZvT is not ZerO's best match up. ZerO was known more for his ZvP. For both players, their games with Fantasy makes up a large portion of their ZvT record (14 games for ZerO and 7 games for Soulkey). For Soulkey in particular, his dominance against Fantasy is one of the first things that are brought up when people discuss his predator-prey relationships with other players. ZerO vs Terran in standard leagues: 57W 51L (52.78%) ZerO vs Non-Fantasy Terran in standard leagues: 52W 42L (55.3%) Soulkey vs Terran in standard leagues: 36W 31L (52.73%) Soulkey vs Non-Fantasy Terran in standard leagues: 30W 30L (50%) As far as the Korean comments, I am Korean so I am just speaking about discussions with other Koreans plus my perception of how each player was pumped up to the viewer by the Korean commentators (rather than a direct comparison). Finally, speaking as a professional mathematician, I think that it is strange for 1) these number of games to be considered a large sample for statistical evaluation and 2) for a 5~7% difference in winning percentages to be considered meaningful over 100 games. However, this is also why I am willing to accept that I could be wrong and Soulkey could have been just unlucky in individual leagues. EDIT: One area of disagreement: I don't think it is fair to describe ZerO's results in individual leagues as merely decent. PS: I will say that I am a little biased because ZerO was my favorite Zerg at the end of BW. Even though Jaedong was obviously the best Zerg by far, I just liked ZerO more for reasons I can't really explain. Yes, I did actually did previously consider perception as of primary importance, in addition to accurate memory, and likely should have mentioned it as well. And yes judgement of quality would be subjective to some extent, even if nevertheless accurate. Specifically the item of accurate "objective" perception (probably of both quantitative/qualitative factors) is likely the primary factor at this point, based on the progression of the discussion. (We can set aside quality for the most part, I will however, add one relevant point: some particular aspects regarding game quality, while not being especially concrete, are not necessarily completely subjective either: For example, does the player control the game from start to finish (e.g. better than other players would be able to in the same situations), or was there the potential or actual occurence of counterplay from the opponent. How fast/accurate were certain executions in situations where the optimal response is known, etc. There may be better examples actually. But anyways, in this type of assessments, over a number of games, there are a considerable number of data (per game), and I would say, in many of these such aspects I noticed Soulkey had a distinct edge. Obviously perception will still apply to some extent - specific significance questionable.) Regarding the sample size: yes, from a general statistical standpoint, the number of games in question, is not exceedingly large. But in this specific circumstance (e.g. BW, the time period/players in question) they may retain meaningful value. 1) value of statistics regarding Fantasy: Having a relatively large sample size against Fantasy is not necessarily bad in a sense, and retains some measure of value. 1) Fantasy is one of the best Terrans overall, and has pretty good ZvT results, including good performances against a high level Zerg like Jaedong. Based on this fact, a very good showing against Fantasy - even in a smallish ~7 game sample is indicative of very good ZvT ability (possibly, but unlikely to be completely specific to playing Fantasy) - even in complete isolation. ie. the expectation that any zerg, nevermind one who is not especially skilled at ZvT, would beat Fantasy 6-1 in a 7 game series, is essentially 0. Sure the large sample size vs. one player is not necessarily optimal, but in this specific situation (i.e. who was better player/ZvT) I don't know if is necessarily too damaging. 2) regarding Soulkey's ZvT winrate: I think it would be agreeable to say that Soulkey was a comparatively inexperienced player, who was showing relatively dramatic and continuous improvement over time until 2012, so his overall (ZvT) results are less meaningful than for someone like Zero (who has a lot more experience, and who's ZvT winrate stats have been relatively consistent) - Soulkey overall ZvT winrate is something like 53%, but if you look at his last 33 games (arbitrary sure) Record: 22 wins - 11 losses (66.67%). I think this is really more of a relevant indication of his ability, than Fantasy/non-fantasy distinction.I don't think it's necessarily reasonable to say that the 66% is a statistical outlier (perhaps qualitative input would be required, or perhaps not), or that fantasy/non-fantasy is necessarily very meaningful in comparison. Hence, why I originally thought it would be more to the point to go down different lines of discussion. 3) Regarding the significance of 5-7% winrate differentials: Just to establish some context: The 5% for some reason seems to have significance in BW - e.g. difference between B - A - S class player. So theoretically if there is a objective 5-7%+ differential in skill;- I am assuming it is agreeable to say that this would be indicative of a significant difference in skill (e.g. potentially even B, A, or S, depending on the situation). Anyways, for the same reasons mentioned before e.g. the possibility of a player attaining the winrate statistics that Soulkey achieved, but not having the requisite skills seems fairly low percentagewise. Also consider that while Soulkey is not a particularly an experienced player, he is not particularly inexperienced either, especially given the quality of competition he has faced. If anything, a newer player who does not have the requisite skills to achieve/maintain a very high winrate, is probably going to get crunched pretty badly over a 100 game sample, and certainly you would not expect such a player to dramatically improve winrate with time (obviously a factor would be the magnitude of the difference). I guess the other side of the coin is, how often will it be accurate to say that a player who is 5-7%+ better over a 100 game sample, is actually the worse player,.. So how exactly meaningful/accurate are the 100 game winrate statistics specifically in this context (and perhaps in other 100 game samples)? I'm not sure, but my initial feeling was that they generally will be meaningful. Obviously the # of tournament games will be a fraction of 100, whatever the implications. Perhaps if there is no specific comparison, regarding Korean commentary, then this could be put aside. Although I am still interested in knowing what the direct comparison would have yielded, in that time period, if something comes up. If Zero's Individual League accomplishments were not merely decent, I won't argue that point. It's interesting you mention that Zero is your favorite player - I admit I found the initial characterization you posed of Zero and Soulkey and Sea, and the subsequent response pattern slightly notable for some reason I could not immediately identify. But that is not necessarily here or there. I don't think I have any significant personal bias in this case. Soulkey was not originally a notable player to me. But after watching him improve so much - especially vs T, I became a fan - for some reason I like players that have high ZvT skill. Look up Yellow[arnc]s record against every TvZ "specialist" you can think of - he has a positive record against essentially every one. I wouldn't say Soulkey was ever became of my top favorite players (he may have eventually but who knows), but essentially by force of play I came to see him as the best Zerg in the world at one point. I will start with the conclusion: There's no way you can actually prove the claim that Soulkey was a class above ZerO. There's no way that I can prove the opposite. I don't think you disagree based on what I've read---unless you think that 33 games can prove someone's skill in objective terms. What follows is just bar talk. You can make the case that the majority perception and their criteria are wrong. That may all be true. However, you shouldn't be surprised that ZerO was considered an S-class player because most people go by OSL/MSL performance when they rate players. In terms of OSL/MSL performance, Soulkey barely has a leg to stand on, while ZerO has a long list of accomplishments beyond anything Soulkey achieved in those leagues. Again, that's unfair, but you can't be surprised about the perception itself even if you disagree with its conclusion. The reason I chose to go by individual tournament records is because that is what the players themselves care about the most. When SKT made Bisu focus on ProLeague practice he even took the unprecedented step of complaining publicly that it was hurting his individual league performance. I've already indicated that considering individual league performance is not somehow more scientific. It's just that I've chosen to focus on the games that mattered the most. That doesn't make it right, but this is what people do when the numbers don't yield a clear answer. There are also BOx type matches in individual leagues, which requires an interesting extra layer of planning. About Korean commentary, the most specific quote I can remember is the one I included in my earlier post. Kim Carry said that "ZerO is a player who never won a championship, but was worthy of one." That's pretty much the highest compliment you can give someone who did not win either league. My perception comes from statements of that kind often being made about ZerO but never about Soulkey. I noticed your comment about ZerO never being S-class, but he was actually the one being referred to as S-class on TV. Most of this is really just due to the aforementioned Korean bias in favor of individual leagues---which might be unfair, but it is what it is. Sea had a higher peak ELO than both ZerO and Soulkey, but he was never considered S-class precisely because of his individual league performance. Even now Kim Carry will point out that Sea was always missing that extra 2% needed to become a champion. That specific conclusion is not particularly relevant. To clarify: I wasn't necessarily claiming Soulkey was class above Zero, but rather that even a seemingly small difference (assuming it's valid) of 5-7%, would actually be quite significant in terms of relative skill difference - whether it is specifically a class above, was not particularly important. (I did mention that the statement regarding class- was just to establish context) Regarding the sample size, in certain cases, a smaller and/or more recent, sample size can actually be more meaningful, than larger more lengthy ones. Perhaps the relevant one is Soulkey's mediocre ZvT overall record ~53%, in comparison with the most recent run of 22-11 66%. No one can reasonably claim that Soulkey in the last couple years was playing at a 53% ZvT skill level. Sure if he was somehow luckboxing/cheesing/playing bad players, etc. maybe the validity of his skill could be questioned, but given the fact that he was a relatively young, constantly improving player, the recent run of 66%, it's pretty clear that considering the larger, more lengthy sample size was actually less meaningful - in fact, highly inaccurate for the discussion at hand. Perhaps it is true that players care more about Individual League. But think about the context. These players are practicing 12 hours a day, often with Proleague success prioritized over Individual League, and they get to play on televised games very rarely compared to the amount of practice they are putting in. They are going to play at a very high level (or try to) whenever it is an official match. It is not wrong to put some degree more importance on Individual results, but I don't think that in light of all the facts, that Individual results retain so much importance in this case: If anything the most accurate way of determining player skill - is the series of inhouse ranking tournies, etc. This is often the primary measure of skill with regards to those directly involved in the Pro Scene. Consider cases, like Jaehoon - most people think he is a terrible player, but the fact is he regularly did exceptionally well in inhouse ranking tournies - indicating very high skill - which is why the coach often put him in ace match situations, even though his results were very poor officially. Additionally, many players highly respect Jaehoon's ability (mentioned in interviews). This is not even an issue with Soulkey - He is consistently winning more inouse tournies than Zero, Zero thinks Soulkey is better, Soulkey has much better statistics in every matchup ----buut Zero better Invididual League results - It's not really a hard computation - I am not discriminating against "popular opinion," more so the goal is just to get an objective view of who the best player is. Again, even if you want to perhaps correctly prioritize Individual League games, and assign more weight to them - it is still not nearly enough to discount all the other significant factors in play here (regarding who is the better player). I don't speak Korean, but this is the first time I have heard anyone imply that Zero was an S-Class Player. Obviously I don't know the specific TV quote you were referring to: maybe it was just an incidental discussion of S-class and Zero, without a direct assertion that Zero was in fact S-Class (which is completely different). But truly, that would be absolutely ridiculous. I fear the discussion is starting to devolve significantly at this point. Let me get to the point. Even if you want to arbitrarily focus on Starleague results, here is what S-Class level results look like: Jaedong: Winner: 2010 WCG Korea Finals 10 games
Runner-Up: 2010 Korean Air OnGameNet Starleague Season 2 18 games
Runner-Up: 2010 Bigfile MBCGame StarCraft League 19 games
Runner-Up: 2010 Hana Daetoo Securities MBCGame StarCraft League 16 games
Winner: 2009-2010 NATE MBCGame StarCraft League 14 games
Third: 2009 WCG Korea Finals 10 games
Winner: 2009 Bacchus OnGameNet Starleague 12 games
Winner: 2009 Averatec-Intel Classic Special Match 5 games
Winner: 2008-2009 Batoo OnGameNet Starleague 19 games
Winner: 2008 WCG Korea Finals 9 games
Winner: 2008 GOMTV TG-Intel Classic Season 1 17 games
Runner-Up: 2008 Arena MBCGame StarCraft League 15 games
Winner: 2008 GOMTV MBCGame StarCraft League Season 4 16 games
Winner: 2007 EVER OnGameNet Starleague 13 games
Winner: 2007 Seoul International e-Sports Festival 19 games
Refer to also: Flash, Bisu, Stork, etc. Here is what S-class Tourny results don't look like: Zero: Runner-Up: 2011 ABC Mart MBCGame StarCraft League 16 gamesIt's such a ridiculous comparison, I don't even wan't to comment on it much further. But suffice to say, I think that anyone could legitimately be surprised that Zero was supposedly considered S-Class(which really I still doubt specifically happened), even if you only want to only look at Starleague results. Again on the balance of all the information provided (including potentially accurate higher weighting of Starleague games), I think the position that Soulkey was the better player (whether it is specifically a class or more or less is not too important), is much stronger. I think I have provided most of the strongest/relevant arguments already, and I don't really think there are many more relevant/significant points to add (to further the discussion), or that there is the need to do so. LOL. Maybe the difference comes down to the fact that I consider Jaedong God-class, which is beyond S-class. :-) Definitions, definitions... Jaedong is the greatest Zerg of all time...kind of an unfair comparison. Also, let's not boil down Zero's resume to that one line from Liquipedia when I've already given a more complete list. I am not saying that you can't argue for Soulkey. I don't disagree that you're making a case. It's just not an objective logical proof of anything, although you are somehow beginning to convince me of the conclusion regardless. PM me if you want to talk some more. I don't want to keep bumping the thread with what's turning into a two-person discussion. Having this kind of conversation again really brings back memories. Who do I need to thank for this? Sonic or FIX? EDIT: long quote spoilered
To be fair, I referenced a number of other players who would be considered S-Class, (and there are even a large number of players never considered as S-Class, who would have much better Starleague results than Zero as well) and the comparative difference to Zero, would be roughly the same. And a complete list of Starleague results, regarding the players I just referenced, in comparison to a complete list regarding Zero, would additionally yield roughly same difference (i.e. top 3 finishes, or # of round of X finishes, respectively) so I did not see any inconsistency with the data used. Although I would have no problem referencing the complete lists of data for each player - except for the extra effort.
I think with essentially any proof, there are degrees of objectivity - ranging from low to high. Certainly the discussion at hand is not one of the highest objective discussions ever, but it retains enough objectivity for me to feel confident that the conclusion I reached is reasonable and accurate.
Like I said, I'm not sure if there is much more from my end, that I could necessarily provide, to further the position, at the moment. If something additional comes up- maybe fine, if not, also fine.
|
|
|
|
|
TLADT24920 Posts
Good writeup. Bisu's looks getting challenged. Does Bisudagger know about this? lol.
cool. Wonder if they did the same for NA/EU as well hmm
|
|
|
|
|
|
EPT
