MicroBob
11-09-2004, 07:02 PM
I have previously written in threads by players bothered by the 'amazing' long-shot/s that came through against them by comparing poker-stats to baseball stats.
Those who don't know much baseball probably won't understand as much what I'm driving at.
Anyway - if someone gets bothered by an all-in 85-15 situation going against them....I make a comparison to a national-league pitcher getting a base-hit. A decent hitting pitcher might have an average of around .150.
If they are bothered by the same player beating them twice on 85-15 situations then I compare it with a pitcher collecting base-hits in consecutive AB's.
If someone loses a couple of 97%-3% all-in's in the same tournament I compare it to a shortstop's fielding percentage (which I think .970 is kinda typical).
So if you think that a 3-percenter coming through twice is 'proof' that the game is rigged....then you'll have to be standing-up in your seat and shouting like a looney-bird if Derek Jeter happens to commit 2 errors in the same game.
You might consider shouting something like "This game is SOOOOO rigged. It's obvious that the odds against that happening are astronomical. We should all demand our money back."
These comparisons are kinda simplistic obviously....but I think they help drive home the point for those who confuse 'long-shot' with 'should never happen in a million years'.
Anyway, somewhere in my long-drive from Wisconsin to Florida or standing in the shower or I-dont-remember-where I started thinking about win-rates and # of hands played with a baseball team's win-loss record.
Please keep in mind that I only have a vague understanding of variance and SD, etc etc....so if my numbers are WAYYYYYY off that's fine....just driving home the general point that the long-run is typically longer than most new players seem to realize.
Anyway, we frequently inform newcomers that 5k hands is not nearly enough to tell if you are truly beating the game or not and some of these players are surprised to learn that 5k hands is actually 'short-term' in the poker-universe.
There is some dispute as to the number of hands where you can really put some faith in your win-rate (+/- 1BB or whatever).
Some say 30k....maybe 100k.
For purposes of my analogy I decided on 80k hands.
A baseball team plays 162 games in a season.
I'm rounding down the season to 160 games.....and am comparing each individual baseball game with each poker-block of 500 hands.
Thus, if you are talking about your win-rate after 10k hands and thinking it means you are a long-term winner....that's basically the same as a baseball team having a winning record after 20 games and thinking that means they will finish the season with the same kind of winning percentage.
This just doesn't happen....there is typically some hot team that starts 14-6 that doesn't even make the playoffs.
This year the Cincinnati Reds had the best record in both leagues on Memorial Day. I think that's about the 45-game mark.
By the end of the season the Reds were about a million games out and didn't even have a winning-record.
Anyway, the comparison seems to hold-up reasonably well the more I think about it although I don't have any numbers on how many BB/100 = how many games over .500 a team might be. Just looking at it generally.
A winning poker-player having a losing streak over the course of 10k hands is perhaps somewhere around the chances of a winning baseball team having a 7-13 or 8-12 record over a 20-game stretch.
You don't expect the winning-player to lose over 10k hands...but it does happen.
You also don't expect the Yankees to post a losing record over a 20-game stretch.....but in the course of a 162-game baseball season there will likely be some 20-game block in there where they go 7-13 or something.
Those are just my general thoughts on the matter and I thought I would share.
Those who don't know much baseball probably won't understand as much what I'm driving at.
Anyway - if someone gets bothered by an all-in 85-15 situation going against them....I make a comparison to a national-league pitcher getting a base-hit. A decent hitting pitcher might have an average of around .150.
If they are bothered by the same player beating them twice on 85-15 situations then I compare it with a pitcher collecting base-hits in consecutive AB's.
If someone loses a couple of 97%-3% all-in's in the same tournament I compare it to a shortstop's fielding percentage (which I think .970 is kinda typical).
So if you think that a 3-percenter coming through twice is 'proof' that the game is rigged....then you'll have to be standing-up in your seat and shouting like a looney-bird if Derek Jeter happens to commit 2 errors in the same game.
You might consider shouting something like "This game is SOOOOO rigged. It's obvious that the odds against that happening are astronomical. We should all demand our money back."
These comparisons are kinda simplistic obviously....but I think they help drive home the point for those who confuse 'long-shot' with 'should never happen in a million years'.
Anyway, somewhere in my long-drive from Wisconsin to Florida or standing in the shower or I-dont-remember-where I started thinking about win-rates and # of hands played with a baseball team's win-loss record.
Please keep in mind that I only have a vague understanding of variance and SD, etc etc....so if my numbers are WAYYYYYY off that's fine....just driving home the general point that the long-run is typically longer than most new players seem to realize.
Anyway, we frequently inform newcomers that 5k hands is not nearly enough to tell if you are truly beating the game or not and some of these players are surprised to learn that 5k hands is actually 'short-term' in the poker-universe.
There is some dispute as to the number of hands where you can really put some faith in your win-rate (+/- 1BB or whatever).
Some say 30k....maybe 100k.
For purposes of my analogy I decided on 80k hands.
A baseball team plays 162 games in a season.
I'm rounding down the season to 160 games.....and am comparing each individual baseball game with each poker-block of 500 hands.
Thus, if you are talking about your win-rate after 10k hands and thinking it means you are a long-term winner....that's basically the same as a baseball team having a winning record after 20 games and thinking that means they will finish the season with the same kind of winning percentage.
This just doesn't happen....there is typically some hot team that starts 14-6 that doesn't even make the playoffs.
This year the Cincinnati Reds had the best record in both leagues on Memorial Day. I think that's about the 45-game mark.
By the end of the season the Reds were about a million games out and didn't even have a winning-record.
Anyway, the comparison seems to hold-up reasonably well the more I think about it although I don't have any numbers on how many BB/100 = how many games over .500 a team might be. Just looking at it generally.
A winning poker-player having a losing streak over the course of 10k hands is perhaps somewhere around the chances of a winning baseball team having a 7-13 or 8-12 record over a 20-game stretch.
You don't expect the winning-player to lose over 10k hands...but it does happen.
You also don't expect the Yankees to post a losing record over a 20-game stretch.....but in the course of a 162-game baseball season there will likely be some 20-game block in there where they go 7-13 or something.
Those are just my general thoughts on the matter and I thought I would share.