HU Strategy

[RacersEdge]

The (almost) ultimate goal is get to the final 2 players with the biggest stack. I am finding this situation (HU with the bigger stack) difficult to play because I don't know how tight to play. Obviously, the blinds are huge so you can't be real tight, but is there a more detailed basic strategy than just not being ultratight? I admit I do find HU a little boring, so maybe if I had a strategy to play, it would be a little more interesting - and profitable.

I don't see much in TPFAP or any previous threads on this site on HU strategy.)

[ddubois]

With 8000 chips on the table, someone will necessarily have less than 10xBB once the blinds hit 200/400. At that point, anything other than pushing is arguably pointless. However, IMO, if your opponent will fold to minraises more often than he should, you may be able to exploit that for awhile (until he catches on, at which point you might want to change gears and start pushing as the default raise size).

Eastbay has said that at this blind level the best strategy and counter-strategy is to push all-in the top 60% of hands (roughly any pair/ace/king/queen, any two suited or connected cards higher than 8?), and call all-in the top 30% of hands (roughly any pair/ace, most big kings, any suited broadways?).

However, it is my theory that if opponent seems to be raising more tightly than proper according to eastbay's guide, I think your calling standards can tighten up, first off by dropping small aces (because of domination).

[CrisBrown]

Hi Racers,

My general heads-up strategy is this: I want to win every hand where I have or catch something, and every hand where neither of us has or catches anything, leaving my opponent with only those hands where he has or catches something and I don't. Aggression, aggression, aggression.

Cris

[eastbay]

Note that this analysis is for the best strategy in the sense that the smallest edge can be gained against it no matter what your opponent does.

That is very different from "best" in the sense of winning the most against a given opponent's strategy. Putting your opponent on a strategy and adjusting to him/it is where the big equity lies. Not in playing in a static way where you're guaranteed not to be more than a 2% dog (which is what the 70/30 strategy is). But if you're clueless, it's a good starting point. (not suggesting either you or the OP are clueless.)

eastbay

[La Brujita]

One other suggestion is play a few Heads Up sit and gos at stars to improve your game.

I like heads up because you get to play almost every hand.

I am just speculating but perhaps what you find boring is hu with shallow stacks becomes a bit like a crapshoot rather than poker.

HU with deep stacks is pretty interesting.

Just my two cents and good luck.

[RacersEdge]

[ QUOTE ]
Eastbay has said that at this blind level the best strategy and counter-strategy is to push all-in the top 60% of hands (roughly any pair/ace/king/queen, any two suited or connected cards higher than 8?), and call all-in the top 30% of hands (roughly any pair/ace, most big kings, any suited broadways?).


[/ QUOTE ]

Great stuff. Is this algorithm independent of stack size?

I am just speculating but perhaps what you find boring is hu with shallow stacks becomes a bit like a crapshoot rather than poker.


[ QUOTE ]
I am just speculating but perhaps what you find boring is hu with shallow stacks becomes a bit like a crapshoot rather than poker.

HU with deep stacks is pretty interesting.


[/ QUOTE ]

I think you're right.

[eastbay]

[ QUOTE ]
[ QUOTE ]
Eastbay has said that at this blind level the best strategy and counter-strategy is to push all-in the top 60% of hands (roughly any pair/ace/king/queen, any two suited or connected cards higher than 8?), and call all-in the top 30% of hands (roughly any pair/ace, most big kings, any suited broadways?).


[/ QUOTE ]

Great stuff. Is this algorithm independent of stack size?


[/ QUOTE ]

Of course not.

eastbay

[RacersEdge]

OK, let me just throw out a situation:

HU, Blinds = 300/600,

Your stack = 6500
Opponent stack = 1500

You are in the SB with K /images/graemlins/spade.gif 4 /images/graemlins/heart.gif

What's the play?

[rjb03]

Push here he only has 900 to call and K high is likely good.

[rachelwxm]

He would probably correctly call w anything. Even that pushing is correct since K4o is better than average hand and you are getting better than 1:1 to push since you post small blind.
w this blind and stack size, pushing w anytwo is not far off since you are getting 1:1.5 for his money and that covers 90% of possible hands. /images/graemlins/cool.gif

[kleraudio]

how bout this scenario

HU blinds = 300/600

You - 4500
Opponent - 3500

You are in SB and hold K /images/graemlins/spade.gif 4 /images/graemlins/heart.gif

what is your play now??

[rachelwxm]

a typical 70/30 rule calls a push here.
But if your opponent do not defend his blind well and only reraise w absolute mosters, you might try to steal with min cost. HP play is very player/flow dependent.

[DrPhysic]

Here are some heads up or shorthanded references that may help.

Brunson SS NLHE “Short Handed play” P510.
Ciaffone, Improve Your Poker, “Shorthanded Poker”, p36.
Ciaffone, Middle Limit Holdem Poker, “Shorthanded Play”, P306.
Sklansky, Theorey of Poker, ch 21 “Heads Up on the End”, P199 (Only partially applicable to HU play, read in context),
Sklansky & Malmuth, HEPFAP, Part 5, “Playing Short Handed”, p183.
Sklansky, HEP, “Head Up vs Multi-Way”, p70 (this is really about multi-way, you have to invert his logic).
Suzuki, Poker Tournament Strategies, “Playing Short-Handed Poker”, p103.
Sklansky, TPFAP, “Down to 2 Players”, p86.
Malmuth, Poker Essays, “Playing Shorthanded, Part 1&2, p151.
There is a reference in Vorhaus, Killer Poker, but as it is not indexed, I can't seem to find it.

I second the notion of playing HU games at stars starting at lower buyins and moving up as you become comfortable with the game. That is what I have done over several months thanks to some input on 2+2. My short table and HU game has improved considerably from that practice.

Doc

[rachelwxm]

thanks, how is the blinds increase for HP there? And how long does one typically last? I guess once blinds are sizable, it gets closer to normal SNG HPs?

[DrPhysic]

Rachael,

The early play can be very conservative or can end in one hand. A HU game will normally last 15 or 20 minutes as an estimate. You need to read your opponent as quickly as possible. I often throw in a minraise bet just to see if he will fold. I often throw in a more than 4-5bb bet with a half decent hand just to see what he will do. I often call an early 300 bet, just to see what he has. Sometimes I find the player is extremely loose and adjust accordingly. If I throw in that 300 call and find out he had KK rather than a bluff or semi-bluff, I have to tighten quickly.

You are right that as the blinds increase, the game becomes more like the last two at a SNG table, because the stack/blind ratio has become somewhat the same, and you hopefully have a decent read on the other player which should be true HU in a SNG. This starts to occur at 50/100 and is certainly true at 100/200.

Remember in your play that, unlike a full table, he also only has one person to read, therefore, you must learn to change gears quickly.

Doc

[RacersEdge]

That would be the kind of hand I can't see pushing...K4o is a 52% fav against a random hand, so I see that meaining the only calls you are going to get are from someone in that huge 48% zone that has you beat.

[rjb03]

[ QUOTE ]
That would be the kind of hand I can't see pushing...K4o is a 52% fav against a random hand, so I see that meaining the only calls you are going to get are from someone in that huge 48% zone that has you beat.

[/ QUOTE ]

Good luck waiting for a hand that is a big favorite against random hands at these blinds. Also, who said you want a call with K4o? Picking up the blinds here would be great and would happen much of the time. If you'll only get called when you're beat as you say, I would push much more often.

[TacoVendor]

I won a PL SnG last night with K3o as the last hand. It beat out QJo when neither of us paired up anything.

My stack at that point was at about 6500 and the blinds were at 400-800. I had posted in the SB and raised and he pretty much had to call with anything there.

We actually went heads up pretty early in the blind structure at Party so I was able to portray a very tight image early on by folding a few times in the SB (blinds were 100-200 at that point). I honestly had junk cards early on but was folding off the SB and just completing and then folding to a raise preflop. After doing this a few times and then us getting right about even in chips (he was less than 500 up on me) I took advantage of the blind increase and started pushing hard.

I find it typical at the Party low levels ($6 and $11) that once someone gets a 'read' about your play they don't notice quick enough when it gets changed up. With the way the blinds increase so fast in HU play you can take out a good portion of their stack before they accomodate the change in strategy.

[rachelwxm]

I did not advocate pushing 100% of time here. If you can steal cheaply, by all means. But I would argue pushing is not that bad. Let’s use a little bit math, shall we?

We know K4o is 1.1 against any two, 1:2.1 dog against top 25%.
According to ICM your equity
Before posting blinds 41.3
Push K4o and no call 42.8
Push K4o and lost all in 32.5
Push K4 and win 50

So if your opponent will call with top 25% of hands, your average equity is 41.6 and if he call with any two, your average equity is 41.7. Both are higher than 41.3!

This means it’s at least a slightly positive EV to push. The tigher your opponent the more positive EV push is.

If you feel you can significantly outplay your opponent, then you might want to wait for better odds. Simple strategy can only mitigate skill gap between you and expert player, not make you an expert player. The latter requires a lot of hard work.

Just my 2 cents.
/images/graemlins/cool.gif

[eastbay]

[ QUOTE ]
I did not advocate pushing 100% of time here. If you can steal cheaply, by all means. But I would argue pushing is not that bad. Let’s use a little bit math, shall we?

We know K4o is 1.1 against any two, 1:2.1 dog against top 25%.
According to ICM your equity


[/ QUOTE ]

Just a little nit: you don't need ICM for heads-up. Prob. of 1st is fraction of chips, and there's only one other possibility.

eastbay

[RacersEdge]

What's ICM - some kind of Monte Carlo simulator?

Anyway, I calculated that pushing has a positive chip EV. Assuming your pushes get called 30% of the time, it's like a 675 chip EV.

The thing I'm thinking of is the variance - in this case about a 15% chance your chips go from 4500 to 1300. If you can have a strategy that trades EV for downside risk, it might be more optimal - and that strategy obviously is a function of the style of play of your opponent.

[rachelwxm]

Indenpendent chip model.
I agree positive EV with a huge variance here, but if your opponent start to pushing hands like K3o, not calling it create a big disavantage for you.

[rachelwxm]

Thanks. and simply because my ICM calculator is so easy to use does not mean that there is no easy calculation. Good point! /images/graemlins/cool.gif

[AA suited]

[ QUOTE ]
Note that this analysis is for the best strategy in the sense that the smallest edge can be gained against it no matter what your opponent does.

That is very different from "best" in the sense of winning the most against a given opponent's strategy. Putting your opponent on a strategy and adjusting to him/it is where the big equity lies. Not in playing in a static way where you're guaranteed not to be more than a 2% dog (which is what the 70/30 strategy is). But if you're clueless, it's a good starting point. (not suggesting either you or the OP are clueless.)

eastbay

[/ QUOTE ]

what 70/30 strategy??? you mean push with top 70% hands and call top 30%?