Do the math: Poker Math

[TheGeneral+]

Hold-em.

A/K vs A/Q on the draw.

What is the chance % that A/K will win and how do you get a percentage?

Thanks

 

[lander]

This is tough. The AK is looking to draw a straight, but if the Q comes on the turn then you're in deep.

I have no idea - I suck at poker.

90%?

 

[valueman]

A/Q has big problems basically needing a queen which there are 3 from 48 cards - five goes and thats IF a king doesn't show. The straight possibilty really doesn't make that much difference as there is only one straight either pair can have. I'd sat A/Q has about a 21% chance approx - would love to know what it exactly is if there are any poker brains out there.

 

[GreenDoberman]

it could be figured out but off the top of my head only three cards help AQ on flop, that is the other three queens, where the other 49 cards keep the AK a favorite, so I'd have to say going into the flop the AK is a 49-3 or 16.3-1 favorite.
That would be about 95 or 96 % correct?

 

[lander]

Green,
K, J, 10 (no Q)

 

[valueman]

i wouldn't have thought it was that high given that there's about a 6% chance each card for a Queen - thats about 30% and thats if a king doesn't show (70%) approx hence my estimate 0.3* 0.7 = .21 the chance for A/Q hence A/K has a 79% chance.

 

[DarrylParsons]

These are very complex calculations because there are many combinations of flushes, straights, and draws as well (2 pair on the board for example). Usually the calculations are done via simulation. I used to have a program like that from Mike Caro -- "Poker Probe" I think it's called, but somehow it got erased and I lost the disk. But if there is $100 in the pot, then I think the AK has an EV of about $87 is my memory serves me correctly.

[This message was edited by Darryl Parsons on October 14, 2003 at 03:22 PM.]

 

[TheGeneral+]

I seen this in the 2003 tourney and the chance % was displayed. I cannot remember but i think Jason Lester had a A/K and Amir Vahedi had a A/Q. I also think the chance % shown was 97%. I was interested on how that number was calculated.

 

[sammythesage]

Cannot figure proper odds boys

until you know if a/k suited or not and same w/a/q or if k/q same suit.

 

[lander]

Good point Sammy.

The flush factor.

 

[valueman]

yes it does depend on suits etc but 97% - wow. I'd like to here it from a pro though and how its worked out.

 

[TheGeneral+]

Correct Sammy. Indeed a factor. In this example let's say the cards are not suited.

Thanks

 

[DarrylParsons]

Good thing I don't trust my memory too often becuase it sucks. I just found an amazing site that lets you program any hands in for any game, just like the software I was talking about:

http://www.twodimes.net/poker/

Turns out the answer to the question is 75-25

Quite a surprise since this is a classic "dominator hand".

 

[SteveMcGarrett]

Sammy, lets go with A/K unsuited v. A/Q unsuited. General, didn't Vahedi catch his straight on that hand? I think a King came on the flop and thats when the 97% was shown? I don't know, but good question for a thread.

 

[DarrylParsons]

To use the simulator you have to input card values including suits separated by a space. One hand per line, e.g.

As Kd
Ac Qs

Input that and it will spit out the result.

 

[TheGeneral+]

May very well be Steve. My short term memory has gotten the best of me.

 

[valueman]

i wasn't far off at 79%.

 

[TheGeneral+]

The 72% does look right. That may have been what I seen and I am 72% sure it was. I was more curious as to how this is calculated, but that may not be feasible.

 

[DarrylParsons]

The 87% I remembered was for AA vs. AK suited. The worst is AA vs. AK off where the suit of the K matches one of the opponents Aces -- 93.5%. I wonder if it's possible to construct a more lopsided pair than that!? I can't think of one offhand.

 

[lander]

Nice link Daryl.
Thanks.