Actual value of KQs heads-up?

[BadBatsuMaru]

I'm a winning player and I think my endgame is at least as strong as the rest of my play. I'm close to finishing Harrington's 2nd book, and it's got me thinking more about how I play heads-up. I thought I might make myself some charts with what percentiles different hands fall into so I can tweak my play (Harrington addresses heads-up play in terms of where your hand falls, whether it's in the top 10%, 20%, 30%, or 40% vs. a random hand).

As I was thinking about how to make my charts and analyze my play, I noticed something about Harrington's top 10% of hands:

pairs AA-66
suited AK-A8, KQ-KJ
unsuited AK-AT

I'm a little surprised by this ranking of hands, because I'm generally happier getting A9o (which isn't in the top 10% of hands) than I am getting KQs which is actually in the top 5% of hands!

So this made me fire up PokerStove and look at a lot of hands. Harrington's listings seem good. Here are the hands I looked at that were very close to the "top 10%" cutoff:

KQs = 63.400% vs. random hand
66 = 63.285% vs. random hand
A9s = 62.781% vs. random hand
ATo = 62.722% vs. random hand
KJs = 62.567% vs. random hand
A8s = 61.944% vs. random hand
------------------------------------ top 10% cutoff
A9o = 60.773% vs. random hand
KJo = 60.569% vs. random hand
55 = 60.325% vs. random hand

What I find interesting is the number of times specific hands lose to "better" hands.

KQs = 63.400% vs. random hand (top 5%)
A2o = 54.929% vs. random hand (top 40%)

KQs = 45.558% vs. A2o
KQs = 41.701% vs. random hand with an A

I just think it's very interesting that KQs is in the top 5% of hands as far as how well it plays against a random hand. The only aces in the top 5% are ATs and better. Still, KQs is an underdog to any hand with an A.

What I'm really concerned with is how valid a strategy it is to play hands strongly based on their performance vs. a random hand. A lot of people seem to play much too tight in the later stages of a tournament, but then when it gets heads-up or 3-handed, they'll shift gears and be willing to go all the way with any ace or any pair. According to Harrington, when the blinds are huge you should be glad to call an all-in with any top 20% hand, but I'd say that when a player has been too tight in the later stages of the game, you should be pretty scared to call with the top 20% hands (and even the top 5% or 10% hands) which are unpaired and don't have an ace.

It also seems like hands in the A9-A7 range are much more worthy of pushing heads-up than their low ranking implies.

Any ideas? Am I some kind of idiot?

[lastchance]

You should create a range of hands, and learn how Harrington's rankings stack up against ranges that are not random.

You will find hands like KQ go down very quickly in value when you start tightening up the ranges, compared to 66.

Harrington's ranges should be good though.

[jon462]

id still rather have A9 than KQ.. maybe cuz my pushes get called by A3 more frequently than K7..

[BadMongo]

As the other replies have hinted at, the problem with ranking the rands against a random hand is that your opponent is not calling a push with a random hand. This is why intuitively you feel that A9o is better to push with than KQs; it performs better against a typical opponent's calling range. What you need to do is put your opponents on a range, then evaluate each how each hand performs against that range range. My guess is that it will be very similar to Harrington's, but might include a few more aces and/or pairs.

[axeshigh]

I was thinking about pushing A2-A5 also, they probably do worse than a random hand like T7 against a lot of calling ranges.

[Nottom]

KQ is only really bad if your opponent has an Ace (plus the obvious monsters).

A hand like A7 is even worse if you know your opponent has an Ace plus it sucks against middle pairs. It just happens to be in pretty good shape against KQ.

The biggest problem KQ has is that when people raise its much more likely they do have an A, plus when you raise you don't have an A in hand to prevent your opponents from having it like you would with A2, so you get called more.

[ChipLeader]

[ QUOTE ]
id still rather have A9 than KQ.. maybe cuz my pushes get called by A3 more frequently than K7..

[/ QUOTE ]

BINGO!
I think most players would rather play a flop with KQ than A9 because KQ plays better. That is, when you hit a pair it is more often the top pair. With A9 you're hoping to hit your A because the 9 so rarely is a top pair. When talking all in though, A9 is better because it so often dominates in all in SnG decisions. In a MTT, you wont see the bad aces pushing so hard because your M is generally higher. In these cases, QK is better because you are less likely to be dominated. Since this is the one-table forum though, i would agree that since the blinds are so high compared to stack and people are pushing any A, A9 is a much friendlier hand to see.

[BadBatsuMaru]

OK, as long as I'm not crazy. [img]/images/graemlins/smile.gif[/img]

I guess Harrington does actually address these issues by saying any pair is strong (since it's never a big underdog except against a higher pair) and by briefly addressing playability after the flop.

That is, with a pair or a ragged A, you're always pretty happy to push PF because you can have trouble after the flop (if you totally miss, you're not going to want to see the turn/river, and if you have a low pair or hit bottom pair with A5 and your hand is good, you're probably only going to get a call when you're beat), whereas whith a hand like KQ or even JT you're still happy with your hand, you just want to actually take the flop and see how things go from there.

And this is how I play it, it's just nice to try and put it all together so I can understand better why I play it like this. Maybe there is a good way to quantify the playability of a hand (hands that you want to maybe call or put in a smaller raise than you think you should) vs. the pushability of a hand (bad aces, lower pairs) where you want to finish the hand up faster because you'll have trouble playing the flop.

[BadBatsuMaru]

I guess KQs isn't actually in the top 5% of hands.

If you go to select a hand in PokerStove it pops up that "Hold'em Distribution" window, and at the bottom it has a slider where you can select hands based on %, but the slider is incorrect.

According to the slider, top 4.8% =
pairs AA-99
suited AK-AJ, KQ
unsuited AK

But if you actually calculate yourself vs. a random hand, top 5.43% (the top 12 hands, better than 65% vs. random) =
pairs AA-77 (77 is 66.236% vs. random)
suited AK-AJ (AJs is 65.393% vs. random)
unsuited AK (AKo, the lowest, is 65.320% vs. random)

So, I think people who use PokerStove should know the slider is worthless. It doesn't include 88 (69.163%) or 77 (66.236%) in the top 5%, but it does include KQs, which is only 63.400% vs. random.

Or maybe the slider is supposed to be used for something else...

[droidboy]

[ QUOTE ]
I guess KQs isn't actually in the top 5% of hands.

If you go to select a hand in PokerStove it pops up that "Hold'em Distribution" window, and at the bottom it has a slider where you can select hands based on %, but the slider is incorrect.

According to the slider, top 4.8% =
pairs AA-99
suited AK-AJ, KQ
unsuited AK

But if you actually calculate yourself vs. a random hand, top 5.43% (the top 12 hands, better than 65% vs. random) =
pairs AA-77 (77 is 66.236% vs. random)
suited AK-AJ (AJs is 65.393% vs. random)
unsuited AK (AKo, the lowest, is 65.320% vs. random)

So, I think people who use PokerStove should know the slider is worthless. It doesn't include 88 (69.163%) or 77 (66.236%) in the top 5%, but it does include KQs, which is only 63.400% vs. random.

Or maybe the slider is supposed to be used for something else...

[/ QUOTE ]

If I remember correctly, the slider ranks hands by preflop equity in a three way pot versus two random opponents. While the ranking is objective according to that criterion, the selection of ranking criterion is entirely subjective. It most useful for getting a starting point for further refinement.

- Andrew

www.pokerstove.com