New heads up poker game - best strategy?

[holdem2000]

Dude, even assuming the 8-card player has 88+, AKs every time he raises, it's still profitable for the 2-card player to push TT+, AJo+, ATs+, KQo, KTs+, QJs (possibly ATo, A9s, KJo, QTs depending on blind:stack ratio). 2:1 on his money is huge. If the 8-card player is bluffing at all the 2-card player probably starts calling with Axs as well. The 2-card player now plays back nearly 1/8th of the time, on average only a 61-39 dog in the hand even versus the 8-card player's premium hand range (and he's getting huge 2:1 odds that more than overcome that).

[holdem2000]

This thread reminds me of the sucker bet of playing a heads up tourney with stacks of 50 and 1/2 blinds... one player doesn't get to look at his own cards, but in exchange gets 2:1 odds. The player with odds is only a 60-40 dog in the tournament if he just pushes every single hand (I think I'm remembering this right, maybe that was for different stack sizes). Some really inexperienced players may lay you even higher than 2:1 when you can't see your cards, just play it as "oh, I'm so good at reading you I don't have to look at my cards" then don't push quite every time so they don't catch on.

[Mook]

[ QUOTE ]
If the stacks were a bit smaller it might be +EV for him to push blind every hand.

[/ QUOTE ]
The stacks are pretty small ... remember, OP's laying 2:1, so Jim's effective stack is only $90, or 45 BB. I'm going to try to work out the math, but I'm almost positive that Jim could be +EV in this game just by pushing every hand.

Mook

[holdem2000]

The stacks aren't small enough. The 8-card player gets dealt a hand with exactly 2 aces in it 9.78% of the time. If Jim pushes every hand and OP only calls when his starting hand held exactly two aces (though he actually calls with more like 77+, AQs+), then OP still averages a profit of about $3.09/hand even when on the big blind (and about a dollar more per small blind).

In fact, if Jim takes a push every hand preflop strategy then OP wins EV on the BB by just calling with aces as long as the stacks are at least $67 each... since OP can actually call with a lot more hands, the push any two strategy for Jim wont be profitable until the stacks are very small - smaller than I first realized.

Edit: When I say stacks of at least $67 each, I mean OP's stack is at least $67, Jim's can be half of that.

[holdem2000]

[ QUOTE ]
Dude, even assuming the 8-card player has 88+, AKs every time he raises, it's still profitable for the 2-card player to push TT+, AJo+, ATs+, KQo, KTs+, QJs (possibly ATo, A9s, KJo, QTs depending on blind:stack ratio). 2:1 on his money is huge. If the 8-card player is bluffing at all the 2-card player probably starts calling with Axs as well. The 2-card player now plays back nearly 1/8th of the time, on average only a 61-39 dog in the hand even versus the 8-card player's premium hand range (and he's getting huge 2:1 odds that more than overcome that).

[/ QUOTE ]

In light of realizing how often the 8-card player has a hand in the higher end of 88+, AKs (e.g. the best hand he can make is AA a lot more often than 88, because he wont make 88 when he could instead make 99+, which will be maybe close to half the time) I retract my earlier claim. Given the current stacks I think the OP may have an edge, but would probably need to make his normal preflop raises very large to capitalize on it.

[Will The Thrill]

after reading this thread, i still think one of the bigger advantages is knowing those 6 cards that P8 discards. especially post flop. so post flop, you know 21.1% of the cards in play, you opponent only knows 9.6%, especially in cases where say your 8 holes cards include 4 spades and 3 hit the board, the chances he's holding even a flush draw are now very slim, but he still has no way to make an educated guess on it. It's be interesting to have adeck of cards right now, just to deal out a few hands in this style 8-2 and see just what the 8 hands hold to get a feel for it

[holdem2000]

I don't think there is much postflop play at all until the stacks are way deeper.

If the stacks are very very small, both players shove every hand.

When the stacks are slightly larger, P2 still shoves every hand, but P8 on occasion folds (not very often as his average hand is around AK or TT, and the blinds are still significant compared to his stack).

When the stacks are a little bit bigger P2 can't push every hand because he's against monster pairs so often and the hands he wins the blinds aren't enough to compensate. Above this point is where the game gets non-trivial.

When the stacks are just slightly into the non-trivial range I think P2 has to switch from his push every hand strategy to a surprisingly tight strategy... any pot he plays he's likely to have P8 push a large pair. P2 might attempt to steal with QQ+ and AJ+ or something like that, the latter predominantly because of card removal effects (these may still be too insignificant to play these hands). P8 will call with something like QQ+ and AK (probably not AK unless P2 plays some unpaired aces), and he will open push a little bit more from the SB, maybe TT+.

Until the stacks get significantly larger I think the game is pretty similar to the last scenario I described. At some point P8 can open for about 1/5th of his stack rather than pushing, but I don't know yet if P2 calls with anything besides the top pairs. P8 probably continuation-bet pushes most flops. P8 probably slightly tightens his opening requirements as the stacks increase (maybe removing TT, AK, and eventually JJ).

Because of the huge value P2 gets from his draws P8 has to avoid serious postflop play until the stacks are very large.

I think this would be a good casino game. The house gets 2 cards, the player gets 8 cards. He posts an initial bet before getting his cards, and after the deal he may either surrender and get half of his bet back, or choose 2 of his cards to run against the dealer's. House pays 1:2 on wins. May need to adjust the starting cards a couple away from 8 to make the house's edge what it would be in a normal table game.

[dfan]

[ QUOTE ]

curious to hear the results u got

[/ QUOTE ]

Well the first session I was feeling my way toward a profitable strategy. About 2/3rds of the time I had a big pair, say QQ+, and raised preflop from $7-$12. However Jim called some of these and postflop play was very difficult. A couple of mistakes by me put me in the hole pretty quick.

So before the second session I thought about optimal strategies a bit. I realized Jim could not call big preflop raises and that he wouldn't get KK+ often enough to counter me stealing his blinds. Also I decided I would play again with no more than a stack of $100. We played the second session and each hand I either limped, folded, or (mostly) raised to $20-$25. When I was able to limp in, I played for set value alone. I folded crap hands (below 88+). Jim tried various counter measures, but he couldn't find an effective one (mainly because he essentially had to give up the blinds every time I raised pf) and I won back everything I lost the first session + a bit more.

I think the strategy I came up with informally fits well with the math-based strategies derived above (My strategy was very similar to Holdem2000's midsize stack strategy. I think his analysis is spot-on about how optimal strategies change as stack sizes change). And I'm pretty sure my side had the advantage when I was allowed to play with a 50bb stack. With a 100bb stack, I still don't know who has the advantage.

[dfan]

Follow-up question:
You are getting the 8 cards. If the following is your hand, which two cards do you play?

A,A,A,Q,Q,x,x,x

[Lord_Strife]

I think QQ is better there because it plays virtually teh same as AA but you have one less set out... it's likely he has a king though so it's tough to see which one is most +EV