|
On December 28 2012 04:45 Cel.erity wrote:Show nested quote +On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly".
No.
You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely.
|
On December 28 2012 04:45 Cel.erity wrote:Show nested quote +On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly".
Every single combination has the same probability. You are comparing 4 straight wins to end the match with ACE game(one single combination) with any other single possible combination. But most other single combination will never be noticed by viewers, they are basically viewed together as one bunch of possible common combinations.
Comparing the single reverse all kill combination with one bunch of seemingly common combinations(but theoretically uncommon for any single one of them), that will make the reverse all kill to be a small probability event. It is just because those two events are not composed of the same number of possible combinations.
|
On December 28 2012 04:51 Talin wrote:Show nested quote +On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely.
This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa.
It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times.
|
On December 28 2012 04:58 Cel.erity wrote:Show nested quote +On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times.
This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T.
The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win.
|
Or maybe it has something to do with them playing 3 of their top 4 players last? I know, call me crazy!
|
On December 28 2012 05:02 Talin wrote:Show nested quote +On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win.
Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. I'm telling you that there is LITERALLY NO DIFFERENCE other than the order in which the games are played. If HerO and TaeJa had played their games first for EGTL, then EGTL would have started out 2-0 given this exact series. It's a 6-game set with an ace match, not a 7-game set, so no matter how you scramble the order of the first 6 games, it will always end in an ace match. You're making yourself look very foolish by arguing this point, especially with someone who calculates probabilities for a living.
On December 28 2012 05:09 L3gendary wrote:
Or maybe it has something to do with them playing 3 of their top 4 players last? I know, call me crazy!
Yes, thank you.
|
On December 28 2012 05:13 Cel.erity wrote:Show nested quote +On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way.
That's not what I'm arguing at all. -_-
When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3).
I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about.
|
I gotta say, Thorzain's been disappointing of late. Something's off with him, he's making easy mistakes that just don't make much sense. Looking for him to pick up it up while he's still getting weekly roster spots, as Thorzain's one of my favorite players!
Maybe EG-TL can go 4-1 vs STX Soul!
|
On December 28 2012 05:13 Cel.erity wrote:Show nested quote +On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:On December 26 2012 19:22 Talin wrote:
They're concentrating on Proleague right now and have what many people would consider an experienced and decent roster in Korea (made out mostly of Koreans) doing nothing else but Proleague. This may or may not be the case come spring and summer, but it is the case right now.
As for the close losses, they were much less close than going from 0-3 to 4-3, which is pretty much the luckiest win you can get regardless of lineups. Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. I'm telling you that there is LITERALLY NO DIFFERENCE other than the order in which the games are played. If HerO and TaeJa had played their games first for EGTL, then EGTL would have started out 2-0 given this exact series. It's a 6-game set with an ace match, not a 7-game set, so no matter how you scramble the order of the first 6 games, it will always end in an ace match. You're making yourself look very foolish by arguing this point, especially with someone who calculates probabilities for a living. Show nested quote +On December 28 2012 05:09 L3gendary wrote:
Or maybe it has something to do with them playing 3 of their top 4 players last? I know, call me crazy! Yes, thank you.
You're talking about high school mathematics when you should simply acknowledge what he's referring to, which is a completely valid point.
|
On December 28 2012 05:19 Talin wrote:Show nested quote +On December 28 2012 05:13 Cel.erity wrote:On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:
[quote]
Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. That's not what I'm arguing at all. -_- When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3). I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about.
Yes, winning from being down 0-3 is extremely unlikely, but you claimed that their reverse all-kill was "closer" or "luckier" than some other type of 4-3 which simply is untrue. It's just a matter of which maps are played first. Your first sentence nailed it. Every type of 4-3 is equally lucky, so I don't think you can criticize a team for any order in which they receive their wins.
If anything, moving away from probability and more into sports psychology, the team which performs better later in the match is usually the stronger team overall, so I would argue that a team who reverse all-kills has more determination than a team who lets their 3-0 lead slip away.
|
On December 28 2012 05:28 Cel.erity wrote:Show nested quote +On December 28 2012 05:19 Talin wrote:On December 28 2012 05:13 Cel.erity wrote:On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:
[quote]
It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. That's not what I'm arguing at all. -_- When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3). I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about. Yes, winning from being down 0-3 is extremely unlikely, but you claimed that their reverse all-kill was "closer" or "luckier" than some other type of 4-3 which simply is untrue. It's just a matter of which maps are played first. Your first sentence nailed it. Every type of 4-3 is equally lucky, so I don't think you can criticize a team for any order in which they receive their wins. If anything, moving away from probability and more into sports psychology, the team which performs better later in the match is usually the stronger team overall, so I would argue that a team who reverse all-kills has more determination than a team who lets their 3-0 lead slip away.
My prior post might be wrong then, but my interpretation was simply that a 4-3 ending in EGTLs favor was significantly less likely if the first 3 matches went to T8 rather than if the first three matches had any other outcome.
|
On December 28 2012 05:19 Talin wrote:Show nested quote +On December 28 2012 05:13 Cel.erity wrote:On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:On December 27 2012 11:03 babylon wrote:
[quote]
Not sure how going from 0-3 to 4-3 can be spun as a lucky win? Maybe if all the last four wins were BO or cheese wins? It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. That's not what I'm arguing at all. -_- When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3). I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about.
At this point I would just accept that the other guy can't wrap his head around certain kinds of probability. Statically flipping four heads in a row is much harder to get than getting four heads out of seven flips even if the end result is 4-3
|
On December 28 2012 05:37 henzi wrote:Show nested quote +On December 28 2012 05:19 Talin wrote:On December 28 2012 05:13 Cel.erity wrote:On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:
[quote]
It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. That's not what I'm arguing at all. -_- When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3). I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about. At this point I would just accept that the other guy can't wrap his head around certain kinds of probability. Statically flipping four heads in a row is much harder to get than getting four heads out of seven flips even if the end result is 4-3
The original statement is that a reverse all-kill is luckier than any other type of 4-3. That is what I'm disputing.
|
On December 28 2012 05:28 Cel.erity wrote:Show nested quote +On December 28 2012 05:19 Talin wrote:On December 28 2012 05:13 Cel.erity wrote:On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:On December 28 2012 04:03 Talin wrote:
[quote]
It's not spun, it's basic probability. Every player has at the very least a 20-30% chance to win a game at progamer level. When a team is up 3-0, winning 4 sets in a row to bring it back is the least likely outcome, regardless of how the games went. Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. That's not what I'm arguing at all. -_- When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3). I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about. Yes, winning from being down 0-3 is extremely unlikely, but you claimed that their reverse all-kill was "closer" or "luckier" than some other type of 4-3 which simply is untrue. It's just a matter of which maps are played first. Your first sentence nailed it. Every type of 4-3 is equally lucky, so I don't think you can criticize a team for any order in which they receive their wins. If anything, moving away from probability and more into sports psychology, the team which performs better later in the match is usually the stronger team overall, so I would argue that a team who reverse all-kills has more determination than a team who lets their 3-0 lead slip away.
I claimed that going from 0-3 to 4-3 was the least likely outcome, which was indeed phrased poorly as I didn't think I'd be arguing this. The reasoning was that winning when behind 0-3 is the least likely outcome (winning = 4-3, hence the misphrase).
Every type of 4-3 is equally lucky to any other result before any games have been played. However, the order of wins definitely influences how "lucky" any win is. With alternating wins, the ranges of favorable outcomes (and thus the chance to win) for either team remain as similar as possible until the very end. If one side manages to get ahead, the odds of them being the first to score a 4th win snowball as well. Winning after being down 0-3 is less likely than if you're losing 1-2 after the same amount of games.
As for the psychology, being determined and being lucky aren't mutually exclusive. Determination may make you not crumble under pressure, but you still have to overcome difficult odds (ie get lucky).
|
I don't get it -- if a team wins 4-0 were they "lucky" because that's an unlikely result?
|
On December 28 2012 06:03 Kentredenite wrote:
I don't get it -- if a team wins 4-0 were they "lucky" because that's an unlikely result?
Already addressed and not really relevant either....
|
On December 28 2012 06:03 Kentredenite wrote:
I don't get it -- if a team wins 4-0 were they "lucky" because that's an unlikely result?
Assuming all games are won or lost with a probability of 0.5.
The probability of 4-0 is 0.5^4=0.0625, 8 times larger than probability of a reverse all kill(0.5^7), so theoretically they are not at the same level of luckiness.
So the statement that reverse all kill is lucky, but 4-0 is not, is possibly valid.
Back to my previous statement, for all 7 games, the event 4-0 gets 8 possible combinations, but reverse all kill event only has one possible combination.
I'm not saying EG-TL is lucky, I'm just saying probability is not the way to argue about this point.
|
On December 28 2012 05:52 Talin wrote:Show nested quote +On December 28 2012 05:28 Cel.erity wrote:On December 28 2012 05:19 Talin wrote:On December 28 2012 05:13 Cel.erity wrote:On December 28 2012 05:02 Talin wrote:On December 28 2012 04:58 Cel.erity wrote:On December 28 2012 04:51 Talin wrote:On December 28 2012 04:45 Cel.erity wrote:On December 28 2012 04:41 Talin wrote:On December 28 2012 04:17 Cel.erity wrote:
[quote]
Actually, basic probability would tell you that there's no difference between starting down 0-3 and then going 3-0, or alternating wins and losses until it ends up 3-3. Neither is luckier or closer than the other. Actually, not exactly. That would only be the case if the match didn't end at 4th win for either side and instead continued infinitely. With a goal of 4 wins, at a result of 3-0 the team in the lead must win one out of four games, whereas the team behind must win four out of four games. If your theory were correct, we would have seen the reverse all-kill scenario as often as we see any other result in a BO7. In reality, it's by far the least common one. No. Any type of result among 7 games (not factoring player skill, of course) has an equal chance of occurring. This could mean L,L,L,W,W,W,W, or it could mean W,L,W,L,W,L,L, or it could mean L,W,W,L,L,W,W. Reverse all-kill only seems like the least likely scenario because it's so drastic, and nobody makes a big deal when the wins and losses alternate "randomly". No. You only need to look at recorded data to realize that you are wrong - reverse all-kill doesn't seem least likely because it's "so drastic", it seems least likely because it happens extremely rarely. And it happens so rarely because it's unlikely. This may be due to the psychology of the players when they are down 0-3, I certainly wouldn't argue with that. However, from a "basic probability" standpoint, there is literally zero difference between a match going W,L,W,L,W,L,W and L,L,L,W,W,W,W aside from the order in which the maps are played. If EGTL had played its last 3 sets first, then it starts with 3 wins and ends with 3 losses instead of vice versa. It's seriously high school level mathematics here. When you flip a coin 7 times, there are a finite number of outcomes, all of them equally likely, including heads coming up all 7 times. This isn't the case of flipping a coin 7 times at once - it's a case of flipping coins only until you get 4 H or 4 T. The first time you flip one, it lands on a T, automatically making it more likely for T to "win". Each subsequent flip alters the probability of either side to win. Look man. Your argument is that winning L,L,L,W,W,W,W is "luckier" or "more unlikely" from a probability standpoint than winning 4-3 any other way. That's not what I'm arguing at all. -_- When no games have been played, all the outcomes are equally possible. Every time a game is lost, the number of possible outcomes that favors your team is reduced. In a 3-0 situation, there is only one possible outcome that makes team B win (3-4), while there are four possible outcomes that make team A win (4-0,4-1,4-2,4-3). I can't argue this shit in English, but somebody that calculates probability for a living should certainly understand what I'm talking about. Yes, winning from being down 0-3 is extremely unlikely, but you claimed that their reverse all-kill was "closer" or "luckier" than some other type of 4-3 which simply is untrue. It's just a matter of which maps are played first. Your first sentence nailed it. Every type of 4-3 is equally lucky, so I don't think you can criticize a team for any order in which they receive their wins. If anything, moving away from probability and more into sports psychology, the team which performs better later in the match is usually the stronger team overall, so I would argue that a team who reverse all-kills has more determination than a team who lets their 3-0 lead slip away. I claimed that going from 0-3 to 4-3 was the least likely outcome, which was indeed phrased poorly as I didn't think I'd be arguing this. The reasoning was that winning when behind 0-3 is the least likely outcome (winning = 4-3, hence the misphrase). Every type of 4-3 is equally lucky to any other result before any games have been played. However, the order of wins definitely influences how "lucky" any win is. With alternating wins, the ranges of favorable outcomes (and thus the chance to win) for either team remain as similar as possible until the very end. If one side manages to get ahead, the odds of them being the first to score a 4th win snowball as well. Winning after being down 0-3 is less likely than if you're losing 1-2 after the same amount of games. As for the psychology, being determined and being lucky aren't mutually exclusive. Determination may make you not crumble under pressure, but you still have to overcome difficult odds (ie get lucky).
Fair enough then, just a misunderstanding. I felt like the implication was that EGTL was lucky to win their match at all. Coming back from 0-3 is obviously difficult, but I still don't feel like it's all that lucky, depending on the order of your star players. EGTL is practically starting down 0-2 in every match because they don't have a very deep roster at the moment.
Keep in mind though, if you're gambling on the outcome of a match and you expect one team to win 4 of their matchups, it shouldn't matter too terribly much to you whether your predicted 4 wins come at the beginning or the end of the match. If somehow EGTL could have played in the order of TaeJa, HerO, JYP, TaeJa, Jaedong, Thorzain, PuMa (ace match in the middle), it's very likely that the match would have been a 4-0 or 4-1 for EGTL without their losses even being played. That's why I was arguing that it really doesn't matter where your wins come from as long as you get 4 of them.
|
On December 26 2012 08:17 Serinox wrote:Show nested quote +On December 26 2012 07:46 SenorChang wrote:On December 26 2012 07:21 Serinox wrote:On December 26 2012 05:27 Liquid`NonY wrote:On December 26 2012 01:35 Serinox wrote:On December 26 2012 01:17 Liquid`NonY wrote:On December 25 2012 14:48 AgentW wrote:On December 25 2012 14:47 proofy wrote:On December 25 2012 14:46 Grettin wrote:On December 25 2012 14:45 Serinox wrote:
Khaldor tell Hwanni how to manage a team please. Cough estro cough. I really doubt its hwanni that is the problem here.. How's Estro doing these days? I dunno, ask NonY. It was lack of funds, not personnel, that hurt estro the most and that wasn't Hwanni's fault at all. Hwanni was an excellent manager. I don't know why Serinox implying that Hwanni is a bad manager should concern anyone. Who is Serinox anyway and how does he think he knows something? It's probably a vacuous judgment. Ignore it until something substantive is said. It was a joke lol, get some humor. You are in the wrong place. And this is supposed to tell me what? That you're in the wrong place? And why?
Maybe because you said something very rude about another person in a way that didn't come off like a joke at all, and then played it off as a joke?
|
On December 28 2012 06:08 Cel.erity wrote:
Keep in mind though, if you're gambling on the outcome of a match and you expect one team to win 4 of their matchups, it shouldn't matter too terribly much to you whether your predicted 4 wins come at the beginning or the end of the match. If somehow EGTL could have played in the order of TaeJa, HerO, JYP, TaeJa, Jaedong, Thorzain, PuMa (ace match in the middle), it's very likely that the match would have been a 4-0 or 4-1 for EGTL without their losses even being played. That's why I was arguing that it really doesn't matter where your wins come from as long as you get 4 of them.
Yeah, if HerO and Taeja played earlier, the result could have easily been 4-0 or 4-1 indeed. When I said lucky, I was thinking about having to overcome very difficult odds to get there, not "lucky" as in being lucky to beat a better team (which I don't think Team 8 is).
I think I originally made the statement because someone was referring to the games that EGTL lost as "close losses" implying that they were unlucky to lose them, whereas Team 8 was more unlucky to lose this one than EGTL was in their own 3-4 losses (despite having to play the strongest players last).
|
|
|
|
|
|
EPT
