A True Coin Flip

[dfan]

Text results appended to pokerstove.txt

990 games 0.005 secs 198,000 games/sec

Board: 4d 8h 7d
Dead:

equity win tie pots won pots tied
Hand 0: 50.000% 50.00% 00.00% 495 0.00 { Tc9c }
Hand 1: 50.000% 50.00% 00.00% 495 0.00 { Ah8s }

[hra146]

this is just todays play, right?

[winterGT]

nice

[SmokeyRidesAgain]

More of a coin flip than actually flipping a coin!

[AKQJ10]

It looks more like a small sample. Why just 990 boards?

Two Dimes figures 1.7M boards which I think may be exhaustive (too lazy to confirm). A8o is a 52% favorite here. I'd go with that, not a much smaller sample.

[belloc]

[ QUOTE ]
It looks more like a small sample. Why just 990 boards?

Two Dimes figures 1.7M boards which I think may be exhaustive (too lazy to confirm). A8o is a 52% favorite here. I'd go with that, not a much smaller sample.

[/ QUOTE ]

Didn't you forget the [4d 8h 7d] board in your run?

[belloc]

[ QUOTE ]
It looks more like a small sample. Why just 990 boards?

[/ QUOTE ]

Oh, and the 990 trials comes from the total number of ways two cards can come with 45 unknown cards and two to come (i.e., 4 cards from the two known hands plus 3 known flop cards leaves 45 unknown cards). Once you've run those 990 possibilities, there are no more possible boards.

Incidentally, this calculation gives the same reason for the probability of catching running "perfect perfect" cards (1 in 990). Say you have AK vs. QQ, and the flop is AAK. The QQ hand needs exactly the two remaining queens to win the pot outright. The probability of one of the two coming on the flop is 2/45. Then the probability of the last remaining queen is 1/44. Multiplied out, that's 1/990.

[AKQJ10]

Oops, sorry. Yes I did.

Same result.

[ADLinden]

A true coin flip.