Heads Up Matches

[JABoyd]

What do you mean by 'big fish in a small pond'. I'm not really following. In your opinion how many heads up matches are required for a good sample?

[jay_shark]

A useful formula is the following :

1/(margin of error)^2 = n

If you want your true win-rate to be within 5 % of your sample win-rate , then we require n=1/(0.05)^2 or n=400 games at 95% confidence .

If you want to be within 1 % , we require n=1/(0.01)^2 , or n = 10,000

[JABoyd]

My math was WAY off...I'm an idiot...

Your math is off...

100 matches with 65% win rate = 65 wins 35 losses
5% rake

__________________________________________________ _

$20 game is a $21 buy in
win $19 and lose $21

65 wins x $19 = $1235
35 losses x $21 = $735
-----------------------
$500 profit / 100 matches
-----------------------
= $5 per match

__________________________________________________ ____

$100 game is a $105 buy in
win $95 and lose $105

65 wins x $95 = $6175
35 losses x $105 = $3675
------------------------
$2500 profit / 100 matches
------------------------
= $25 per match

_________________________________________________

Now I play 3 games at a time which would come to $15 per "match" ("match" here is defined as 3 games at the same time) with a win rate of over 65% according to my records.

So now back to the variance...play 2 "matches" and risk $30 or play one $100 match and risk $25???

[JABoyd]

[ QUOTE ]
A useful formula is the following :

1/(margin of error)^2 = n

If you want your true win-rate to be within 5 % of your sample win-rate , then we require n=1/(0.05)^2 or n=400 games at 95% confidence .

If you want to be within 1 % , we require n=1/(0.01)^2 , or n = 10,000

[/ QUOTE ]

I average about 1500 heads up games a month with an average with a winning percentage of 67.57%. And just for reference the break even point with a 5% rake is 52.50%

[RustyBrooks]

Your conclusion at the end is all wrong, again.

Playing 2 $20s risks $42 to win $10
Playing 1 $100 risks $105 to win $25

Notice anything there? It's the same risk/reward ratio. Which is obvious, because it's the same win/rake ratio, and you're assuming the same win rate.

I hope you are not using math to help you play poker.

[RustyBrooks]

Hint: how many $20s do you have to win to make the same as winning one $100?

Answer: 5, and you'll risk $105 doing so.

[JABoyd]

Thats not my point....my point is "In your opinion is it better to risk that same amount of money playing 5 $20 games or 1 $100 game" ??

[JABoyd]

[ QUOTE ]
Your conclusion at the end is all wrong, again.

Playing 2 $20s risks $42 to win $10
Playing 1 $100 risks $105 to win $25

Notice anything there? It's the same risk/reward ratio. Which is obvious, because it's the same win/rake ratio, and you're assuming the same win rate.

I hope you are not using math to help you play poker.

[/ QUOTE ]

And there is no need to attempt to insult me. Becuase I'm sure I have a much strong math background then you could even dream of. Nice try though!!

[RustyBrooks]

So... do you or do you not take issue with my assessment of your summary?

[RustyBrooks]

[ QUOTE ]

So now back to the variance...play 2 "matches" and risk $30 or play one $100 match and risk $25???


[/ QUOTE ]

So does this sentence make some sort of sense to you? Really?