AK and various late tournament 'tricky' situations

Dorkus Malorkus

Dorkus Malorkus

Jul 12, 2005
Total posts
I've seen a few threads recently about playing AK preflop in various different tournament situations, essentially revolving around the "should I call this big shove?" question. Some general points first:

- AK *is* a big hand in the right situation. The statement "AK is only a drawing hand" is a horribly vague and general statement which is of absolutely no use to anyone and is generally just spouted out by the WPT-brigade when they want to sound smart because Mike Sexton said it once. Don't say it ever - it just makes you sound stupid to anyone who has a clue. ;)

- AK is a horrible hand in other situations and should be folded without a second thought.

How do we differentiate between the two situations? Well, imagine two sit and go bubbles. In the first, you're the big stack and it's folded to you on the button with AK. Of course you raise (you can raise with much worse hands too, obviously). In the second, you're second in chips on the button with 10BBs and two small stacks are sitting in the blinds with 2BBs each. The big stack UTG raises and you fold because anything else is absolutely horrible.

Of course, these are contrived situations in which hopefully the best course of action is blindingly obvious. What about when it's a closer decision? I'm not going to give a specific example, there are a couple of threads floating about with good examples if you need to see them (here and here, for two). Let's just say we're 8-handed late in an MTT and in the money. UTG+2 raises 3BBs, folds to CO who shoves for 25BBs, about the same amount as you have in front of you, though he has you covered. Do you call this with AK? Let's ask some questions...

1) What happens if we fold?
2) What happens if we call and win?
3) What happens if we call and lose?
4) How likely is 2) compared to 3)?

For the brief example above, let's answer them. Get yourself into the habit of quickly asking and answering these questions when you're confronted with a tricky preflop call/fold decision.

1) We fold, we preserve a 25BB stack, which in a standard online tournament is enough to put us in the middle of the pack most likely. More importantly, we still have lots of room for maneuver. We can raise and get people to fold, we can reraise and get people to fold. Bear in mind that in the late stages of MTTs the vast majority of the play occurs preflop, and thus your main consideration should be the freedom you have preflop. Don't overcomplicate things by wondering about things like "Well if I raise and someone calls me then I bet and they call my continuation bet can I fire another barrel on the turn with fold equity?", simply because seeing a turn without all the chips being in in a late tournament situation is so rare. Anyway, the fact we have room for maneuver if we fold here is an argument for folding.

2) We call and win (let's assume UTG+2 folds), we double up to 50BBs, giving us an extremely comfortable stack, but by no means guaranteeing us of a win or even a final table (depending on the exact tourney situation, of course).

3) We call and lose, we're busto.

The fourth question is, of course, the real key. We can't make decisions purely based on "What happens if we do x?", we need to establish how likely it is that 'x' will happen. To answer 4) we need to ask ourselves a new subset of questions...

1) What's the tournament situation? How is this likely to affect the mindset and play of the 'big raiser'?
2) What's our general read on the 'big raiser'?
3) What other action has happened in the hand aside from the huge raise from the 'big raiser'?
4) What, therefore, is the range of hands that the 'big raiser' could be holding?

4) obviously follows on from your thoughts on 1), 2), and 3). There are other considerations to make (relevant stack sizes etc), but to be honest dwelling on these isn't the point of this post. However, let's take a look a these questions...

1) In the example, we're in the money. People generally either get very tight and try to slide up the payout ladder while staying out of trouble, or they're just glad to have made the money and are now happy to be reckless in the hope of building up a big stack. The latter is slightly more common than the former from my experiences, but you need to use your reads on the player to supplement the jusgment you come to here, obviously.

2) Is obvious. Is he a LAG, TAG, donkey, does he like to make big moves with crap hands etc etc.

3) The more action, the scarier the big raise should look. If a thinking player raises three limps though, in many situations that's not necessarily the sign of a huge hand.

Again, the fourth question is key, and follows on from the first three. In the vast majority of hands where we're making a decision with AK preflop, there is actually one, and only one really key question.

"Can he be making this move with AQ?"

Let's look at some numbers from PokerStove

equity win tie pots won pots tied
Hand 0: 39.785% 23.66% 16.13% 10936805 7456889.50 { AcKd }
Hand 1: 60.215% 44.09% 16.13% 20381624 7456889.50 { JJ+, AKs, AKo }

equity win tie pots won pots tied
Hand 0: 50.106% 38.23% 11.88% 25529440 7931059.00 { AcKd }
Hand 1: 49.894% 38.02% 11.88% 25388298 7931059.00 { JJ+, AQs+, AQo+ }

That's quite the difference, right? There are few situations in which we want to be calling a shove with 40% equity, but with over 50% equity and some dead money in the pot (either from limpers/other raisers or even just the blinds if significant enough), we can be happy calling! If the 'big raiser' makes the move with hands worse than AQ, then it's not really a close decision anymore, you either obviously have to call because you're ahead of the range or in some situation you will have to fold because the situation dictates it (as in the second SNG example near the start of this post).

What pairs does the big raiser make this move with? You shouldn't care. The 'pair range' has little impact on your equity.

equity win tie pots won pots tied
Hand 0: 49.175% 38.86% 10.32% 29939770 7951130.50 { AcKd }
Hand 1: 50.825% 40.51% 10.32% 31211649 7951130.50 { TT+, AQs+, AQo+ }

Above, we added AQ to villain's range and we gained over 10% equity. Here, we add TT to the range, and we lose less than 1%. In the limited time you have to act online, you can't afford to ask yourself questions which really don't matter that much at the expense of asking the questions that do matter.

Heck, let's go crazy.

equity win tie pots won pots tied
Hand 0: 47.338% 42.23% 05.11% 67243624 8138704.00 { AcKd }
Hand 1: 52.662% 47.55% 05.11% 75723240 8138704.00 { 22+, AQs+, AQo+ }

Now villain's range includes all pocket pairs and our equity has been impacted by not even 3%. And of course, very few players are making a huge (overbet) move preflop with 22-66 unless the situation dictates it (and if we think our opponent is desperate enough to be shoving 22-66, we must surely correspondingly think that they're desperate enough to be shoving a lot of Ax hands).

Pairs impact our equity so little of course because we're not far off 50/50 if we're against them. This principle applies to many hands - don't bother when assigning ranges to consider hands you're flipping with too much - first concentrate on hands you're significantly ahead of and hands you're trailing.

So next time you're faced with a tricky decision with AK, or indeed any hand, make sure you ask the right questions before making a decision. :)

~ Chris


Mar 1, 2007
Total posts
wow... i hadn't realized the difference between T-T and 2-2 in these kinds of situations was so minimal... thank you! That spares me an awful lot of headache....


Forum Admin
Oct 13, 2006
Total posts
Good post Chris - I struggle with playing AK correctly. Can't get through the whole post right now, will absorb the first part and come back to it.


Feb 19, 2007
Total posts
I like playing AK.. it always works well for me. I fold if I see nothing on the flop tho.


Advanced beginner
Sep 20, 2006
Total posts
So the key, is AQ in his range for this particular situation for this particular player? If so, and assuming some dead money in the pot, we should be calling.

In the situation in my thread, probably not. For whatever reason, I expected AK or TT or up.

Thanks Chris.