On September 16 2012 13:31 TommyP wrote:
International games are meaningless stats because its not the top tier play.
International games are meaningless stats because its not the top tier play.
MLG isn't top tier? IPL4 wasn't top tier?
Forum Index > SC2 General |
Femari
United States2900 Posts
On September 16 2012 13:31 TommyP wrote: Show nested quote + On September 16 2012 08:37 lysergic wrote: You should include the number of games rather than just the percentage in the OP. It's misleading to just say PvZ 60%. Korea PvT: 43-27 (61.4%) ZvT: 22-25 (46.8%) ZvP: 28-42 (40%) International PvT: 259-301 (46.3%) ZvT: 290-278 (51.1%) ZvP: 364-322 (53.1%) ![]() Note: I calculated the # of games for each matchup using the winrate * total games on the international graph; they might not be 100% accurate. These statistics, especially the korean ones, are pretty meaningless. Even 1 game has a significant impact on the winrates. For example, if Jaedong beat MarineKing 2-0 rather than losing 1-2 in GSL, then ZvT would be 50%. International games are meaningless stats because its not the top tier play. MLG isn't top tier? IPL4 wasn't top tier? | ||
Assirra
Belgium4169 Posts
On September 17 2012 04:52 Femari wrote: Show nested quote + On September 16 2012 13:31 TommyP wrote: On September 16 2012 08:37 lysergic wrote: You should include the number of games rather than just the percentage in the OP. It's misleading to just say PvZ 60%. Korea PvT: 43-27 (61.4%) ZvT: 22-25 (46.8%) ZvP: 28-42 (40%) International PvT: 259-301 (46.3%) ZvT: 290-278 (51.1%) ZvP: 364-322 (53.1%) ![]() Note: I calculated the # of games for each matchup using the winrate * total games on the international graph; they might not be 100% accurate. These statistics, especially the korean ones, are pretty meaningless. Even 1 game has a significant impact on the winrates. For example, if Jaedong beat MarineKing 2-0 rather than losing 1-2 in GSL, then ZvT would be 50%. International games are meaningless stats because its not the top tier play. MLG isn't top tier? IPL4 wasn't top tier? And yet GSL is one tier higher. We have never seen so far a whole MLG/IPL of code level S players. | ||
Femari
United States2900 Posts
On September 17 2012 04:58 Assirra wrote: Show nested quote + On September 17 2012 04:52 Femari wrote: On September 16 2012 13:31 TommyP wrote: On September 16 2012 08:37 lysergic wrote: You should include the number of games rather than just the percentage in the OP. It's misleading to just say PvZ 60%. Korea PvT: 43-27 (61.4%) ZvT: 22-25 (46.8%) ZvP: 28-42 (40%) International PvT: 259-301 (46.3%) ZvT: 290-278 (51.1%) ZvP: 364-322 (53.1%) ![]() Note: I calculated the # of games for each matchup using the winrate * total games on the international graph; they might not be 100% accurate. These statistics, especially the korean ones, are pretty meaningless. Even 1 game has a significant impact on the winrates. For example, if Jaedong beat MarineKing 2-0 rather than losing 1-2 in GSL, then ZvT would be 50%. International games are meaningless stats because its not the top tier play. MLG isn't top tier? IPL4 wasn't top tier? And yet GSL is one tier higher. We have never seen so far a whole MLG/IPL of code level S players. Yes but that's like saying a player like Symbol isn't a top tier player because he hasn't won GSL when IM is hogging all the titles. There is definitely top tier competition there, and it even outdoes a lot of the stuff in the Korean TLPD. Aren't weeklys in the K-TLPD? Saying that there's no top tier competition internationally because the marquee competition took place in Korea is simply incorrect. | ||
Reborn8u
United States1761 Posts
| ||
keglu
Poland485 Posts
| ||
hunts
United States2113 Posts
On September 17 2012 02:45 Gladiator333 wrote: Show nested quote + On September 17 2012 02:34 hunts wrote: On September 16 2012 17:12 Huragius wrote: Korean ZvT Statistic: ZvT: 22-25 (46.8%) 47 games lol. It's absolutely worthless statistic due to such a little sample size. But because of the level of play it's still worth more than international statistics where they include random GM/masters players that aren't even sponsored or that are on no name semi pro teams that wouldn't even win 1 game in code B. Can you explain why a small sample size is more relevant than a bigger one? I'm confused. You didn't even read the post you quoted did you? Korean sample is much more meaningful despite the size difference because of the huge skill difference. In the international scene there are tons of games in IPL, MLG, NASL, etc.. that have pros vs semi pros, pros vs ametures, koreans vs foreign ametures/semi pros, makes for skewed data because while some of the same top players from the korean scene might be in international, the bottom skill level in it is significantly lower. The international statistics include games of very varied skill levels, therefore skewing the results. Wheras in korea most of the games are played by players of similar standing thanks to GSL doing a fairly good job of sorting players based on performance. Obviously your response was a troll and most likely flame baiting and not a serious question, but there you go. | ||
aksfjh
United States4853 Posts
On September 17 2012 03:53 ohampatu wrote: I really wish TL would put in a rule to stop posting these. They dont represent balance in any way, shape, or form. You either have a large sample size that has a huge range of skill forms, or a small sample size of the tip top. Neither help in determining balance. Id rather see only KR statistics in 3-6 month intervals, and even then, it still wouldn't represent balance that much. The idea is to take the stats as a part of the larger picture. Roll these results in with ladder representation, qualifier results, individual games, and top 4 results from major tournaments and you have a good idea which race is worth betting on. On September 17 2012 04:27 []Phase[] wrote: Show nested quote + On September 15 2012 23:58 VPVanek wrote: So we're back at terran having a problem with Protoss, and zerg having a problem with terran.. 46% is pretty balanced man... The variations we are seeing atm, both in and outside of korea, are partially due to metagame anyways, like with many of the past statistics. But there are always a bunch of dramaqueens who misread the stats in these kinds of threads. Its pretty balanced people, just a few minor tweaks can be made here and there, but not much. Of course it's metagame. Even unbalanced streaks show up as metagame. The question is if the metagame will correct itself. | ||
Fragile51
Netherlands15767 Posts
On September 17 2012 18:34 aksfjh wrote: Show nested quote + On September 17 2012 03:53 ohampatu wrote: I really wish TL would put in a rule to stop posting these. They dont represent balance in any way, shape, or form. You either have a large sample size that has a huge range of skill forms, or a small sample size of the tip top. Neither help in determining balance. Id rather see only KR statistics in 3-6 month intervals, and even then, it still wouldn't represent balance that much. The idea is to take the stats as a part of the larger picture. Roll these results in with ladder representation, qualifier results, individual games, and top 4 results from major tournaments and you have a good idea which race is worth betting on. Show nested quote + On September 17 2012 04:27 []Phase[] wrote: On September 15 2012 23:58 VPVanek wrote: So we're back at terran having a problem with Protoss, and zerg having a problem with terran.. 46% is pretty balanced man... The variations we are seeing atm, both in and outside of korea, are partially due to metagame anyways, like with many of the past statistics. But there are always a bunch of dramaqueens who misread the stats in these kinds of threads. Its pretty balanced people, just a few minor tweaks can be made here and there, but not much. Of course it's metagame. Even unbalanced streaks show up as metagame. The question is if the metagame will correct itself. The only way to find out if the metagame will correct itself is by giving it a lot of time, not yelling at blizzard to "fix this and that" like many people seem to be doing. I'd say right now the game itself needs to be left alone and the balancing must happen with a bunch of new tournament maps, not by balancing stats and spells and whatnot. We have hots beta for that. | ||
FeyFey
Germany10114 Posts
| ||
Evangelist
1246 Posts
| ||
| ||
![]() StarCraft 2 RotterdaM StarCraft: Brood War![]() TKL ![]() PiGStarcraft521 SteadfastSC ![]() IndyStarCraft ![]() BRAT_OK ![]() ![]() NeuroSwarm ![]() UpATreeSC ![]() ProTech92 Dota 2 Counter-Strike Other Games summit1g16288 Grubby14139 tarik_tv12162 FrodaN1965 Liquid`Hasu351 elazer285 Skadoodle112 Maynarde96 QueenE62 Organizations
StarCraft 2 • davetesta33 StarCraft: Brood War• Hupsaiya ![]() • Catreina ![]() • Kozan • IndyKCrew ![]() • sooper7s • Migwel ![]() • AfreecaTV YouTube • Laughngamez YouTube • intothetv ![]() • LaughNgamezSOOP Dota 2 League of Legends Other Games |
PiGosaur Monday
PiG Sty Festival
Replay Cast
WardiTV Invitational
Code For Giants Cup
SOOP
ShoWTimE vs Clem
Replay Cast
ReBellioN vs HonMonO
The PondCast
WardiTV Invitational
Replay Cast
[ Show More ] Replay Cast
Replay Cast
CranKy Ducklings
[BSL 2025] Weekly
Sparkling Tuna Cup
|
|