How to Play AK in Poker: The Best Drawing Hand

John Racener
A-K looks nice, but it can get you in trouble if you don't understand its true equity.

Ask anyone: they'll tell you ace-king, both suited and unsuited, is a premium Hold'em starting hand. So how come it's such a trouble hand for beginners?

As with all the premium hands, the pots you play with A-K are typically going to be larger than average, forcing you to make more frequent and more difficult decisions.

Before you can attempt to formulate strategy for the hand, you're going to need a firm grasp on the real strength of A-K. The best way to do that is by looking at the true equity of the hand in various scenarios.

How to Play A-K

A-K is a tricky hand, in that the range playing back at it is very wide in its statistical strength.

Daniel Negreanu

It's much more difficult to know where you stand with A-K than with many other hands. There is no worse spot to be in than not knowing whether hitting the flop will be a good or bad thing.

Because A-K is a drawing hand (meaning it needs to hit the board in some way to be more than just ace-high), many players believe it's best played in pots with multiple players. Since it has to hit to be good anyway, they feel it's desirable to get as much money in the pot as they can.

The second school of thought is to play A-K like a premium hand, raising heavily to isolate A-K against a single opponent. Before you go any further with one school or the other, let's take a look at the statistics of A-K against both of these options.

How to Play A-K vs. a Single Opponent

The following hands are included in this comparison: AA, KK, QQ, 66, A-Q offsuit, A-Q suited and 7-8 suited.

It makes no sense to run the numbers against every possible hand, so I've chosen the majority of possible situations: overpair, pair under one of your cards, underpair (dead-end), underpair (unhindered), dominated ace and low suited connector.

(All equity calculations courtesy of PokerStove)

Hand Hand % A K %
A A 87.9 12.1
K K 65.9 34.1
Q Q 53.8 46.2
6 6 52.1 47.9
A Q 28.7 71.3
A Q 24.4 75.6
7 8 39.3 60.7

The first thing to note is this list only gives you a brief glimpse into A-K's equity in this context. The idea wasn't to make a comprehensive list, but to get an idea of where A-K stands pre-flop against one opponent.

The Field

Also note that the numbers change by a few percentage points when you change suits around, as shown with the two A-Q examples.

The average equity of A K for all of these examples is 49%. This might be surprising, considering that it's a top 5 starting hand. Although this number is accurate, it's a good example of how statistics - even accurate statistics - rarely tell the full truth.

For example, for every time your A-K runs into AA, you're going to have multiple run-ins with hands such as A-Q, A-J, and K-Q. Statistically, you're more likely to run into QQ than KK, and there are more nondominated suited-connector hands than the contents of this list combined.

If you factored in all of the possible hands, and the frequency of playing A-K against them, you would see the A-K average win percentage climb to a very profitable level.

How to Play A-K vs. Multiple Opponents

Now to run some equity numbers on how A-K holds up against multiple hands in a single pot. I'm using the same range as I used for the single-opponent calculations, but I will set up a few scenarios.

First, let's start with a direct comparison. We're going to assume that your A K got all-in against seven opponents pre-flop, allowing all hands to see a river (unlikely, I know, but this is exclusively for statistical evaluation):

Hand Hand %
A K 10.6
A A 30.8
K K 8.7
Q Q 8.1
6 6 15.5
A Q 11
7 8 15.2

(You might have noticed that I changed the suits of some cards, and left out the second A-Q example. I had to do this to remove all instances of two players holding the same card at once.)

In this unlikely scenario, A-K is going to win the hand 10% of the time (or 1 in 10). Considering you're only getting 7-1 on your money, this is a -EV scenario. In fact, any scenario that has AA in the mix is going to lose you significant money.

(On a separate note, take a look at AA: a 30% win rate while getting 7-1. It's for this reason players such as Mike Caro believe AA is best played multiway to optimize long-term results.)

If we take a more likely scenario, the numbers will change dramatically. In this scenario we're going to put A-K into a limped pot against the type of hands you'll commonly see all at once. If no one raises, chances are no one is holding AA or KK.

Hand Hand %
A K 19.5
A Q 18.6
A 9 2.7
6 5 12.5
10 10 24.1
3 3 15.2
7 9 7.4

In this more likely scenario, our suited A-K is 20% to win the pot. We're going to win 1 in 5, while getting 7-1 on our money. Now we're back in the black.

What is Your Equity with AK?

As I mentioned earlier, this first article is exclusively aimed at helping you understand your equity with A-K. The numbers in this article are the control to start from - ground zero.

What is your equity with AK?
What is your equity with AK?

These numbers are true equity, which is not to be confused with other forms of odds. These numbers only give you an idea of where to start with a hand like A-K. Your goal is to manipulate the numbers and your opponents into giving you better odds on the hand than the base equity offers.

In short: Hand equity is not always the same as the odds. The odds in play are false, due to lack of knowledge. You don't know 100% what your opponent has, and the same goes for them against you.

This means the odds change based on fold equity (your bluffing latitude) and on the choices you make with the knowledge you have.

You have the ability to choose to play or fold the hand. If you fold every time the A-Q hits the queen, and make the call every time your ace hits against KK, you're going to make far more money than the equity predicts.

Another way to think about it: if every time you play your A-K vs. A-Q, you get it all-in for 1,000BBs, and every time you run your A-K into AA, you lose 10BBs, you've just upset the equity predictions.

Even though the equity prediction is accurate in terms of how often you'll win the hand, if you manipulate the amounts of the wins and losses, you make far more money than the equity would appear to allow. (I know this example is clearly not possible; it's just meant to make the concept easy to understand.)

Odds & Equity are Negotiable

Knowledge and action change odds. If the best and worst basketball teams in the world face off against each other, with the former knowing that they're ridiculous favorites to win, it may well affect their respective play - to the extent even of the lesser team getting the upper hand.


We've all witnessed surprise upsets and underdog victories as a result of this exact scenario. If both teams had gone into the game knowing nothing of their own skill in relation to their opponents' - if both teams believed they were the best - the better team's odds of victory would be almost dead-on accurate.

If you want to blow your noodle with this propensity of odds to defy logic, check out the Monty Hall Problem, which in short goes like this: (wording from Wikipedia)

Suppose you're on a game show, and you're given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say No. 1, and the host, who knows what's behind the doors, opens another door, say No. 3, which has a goat. He then says to you, "Do you want to pick door No. 2?"

Statistically, what should you do?

Unless you're a very sharp logistician or mathematician (or have heard this problem before), you will assume that it doesn't matter. Two doors, one prize = 50% chance either way.

This is, in fact, incorrect. You should always request he open the "other" door. By doing so, you actually have a 66% chance at winning the car. The best way to see this is through a chart with all the possible options outlined on Wikipedia.

More Beginner Strategy Articles:

Please fill the required fields correctly!

Error saving comment!

You need to wait 3 minutes before posting another comment.

Aleksei 2013-06-23 21:01:06

I don't like the idea of calling AK a "drawing" hand. For one thing it's not a speculative hand: It will draw something strong a bit over 1/3 of the time, and that something is usually TPTK; AKo makes straights about 2% of the time, and AKs draws to better than top pair just about 10% of the time.

But more importantly, it's not a hand that's advantageous to play timidly preflop, for two reasons: First, as ranges get narrower AK gets stronger, because the more Ace-heavy a range is the more hands AK dominates. And second, as ranges get tight enough that AK starts losing equity against them (about 10% or so), AK picks up a significant and increasingly large amount of fold equity, because it blocks AA and KK such that even if the only worse hand a villain can raise in a spot is QQ, he has QQ half the time and KK/AA the other half. Ergo, you usually should happily re-raise any time you get raised to preflop, because unless Villain can only have KK/AA in a spot and no worse at all (in which case you make an exploitative fold), you easily have enough fold equity pot equity to raise for value or as a semibluff.

And more importantly, you will not win large pots with AK by seeing the flop with it, outside of fairly loose games. When you flop TPTK in a very tight game, your villains will usually have better or fold.

ream 2013-01-26 16:08:21

to understand monty hall it's enough to make the same experiment with 10 or 1000 doors. Then it makes more sense. If you pick one out of 1000 it is very unlikely to pick the right one but if all other but two doors out of 1000 will be opened and you have a choice to switch it makes sense that the other door is 99.9% the right one.

ream 2013-01-26 16:05:55

Tummon you are wrong. For Monty Hall to work the host can know where the car is. What Alex7 wrote is a bunch of egoistic bull crap.

Tummon 2011-05-16 19:05:46

For the monty hall problem to be accurate, the host can't know which door the car is behind. When you talk chance or odds, we're assuming things are happening randomly (i.e. flipping a coin). If the host knows where the car is, he's not randomly picking a door to open.

Bill Newport 2010-10-19 22:33:34

This Monty Hall problem vexes many people, Ph. D's and others don't see that it's auto to switch.

Mad props to the dude who saw this in two seconds, that's how long it should take.

If there are forty-eight other boxes you should still switch, but you only improve your chances slightly: to about 1 in 47 instead of 1 in 48.

Also props to the dude who did the problem with the intention of not even offering a switch if the "contestant" chose the wrong box (which will occur in 2/3 of the cases).

The problem is based on the premise that Monty ALWAYS shows you a swag box, selected by Monty himself.

If Monty just says "You lose" when you pick the wrong box, and "Wanna swap?" when you pick the right box, then of course the odds are effected.

Mad props also, finally, to anyone with so much time on their hands that they're actually reading this. No wonder the economy is in the toilet!

PS: Bridge journals and web-sites write about this a lot, and even champion bridge, chess and poker players often don't get it, so....

alex7 2009-07-31 12:53:00

I have done this to a enginer friend of mine, i put 15$ in one of tree boxes, he paid me 5$ to play, i said to him to chose one but not to open, then he picked the number 1, i open the number 3 and show that its not there. Should he switch to number 2? He switched to number 2 and open, it was empty. I show that was in the other one and i won 5$. He guessed was bad luck because he had 66%. But what he didnt see is that i am a poker player. Because i knew he already know this problem, i only opened another box because he picked the right one, than was almost a 100% chance that he would switch to another box. I read him, i trapped him, i won 5$, and he still thinking "i had the right odds to play", so he will keep coming back. Just like in poker. But he is my friend so i will not do that again, probably...

Very usefull post, th for taking your time writting.

alex7 2009-07-31 09:23:00

Here goes another, try to invert the triangle moving only two points.
. . .
. .

ps: But the one i was asked was: whe have 50 boxes and 1 have a big prize, you pick one, then the person who knows where the prize is open one box where the prize isnt, now you stay with your box or u switch to one of the another 48?
Theres a lot of s* person on the world that apears on these game shows, there is one in my country that u need to guess the price of a object, the one who gets closer wins, there is only two people competing for the prize each time. so if there is a Ball u think its 27,00$, and the oponnent who acts first say is 7,43 what do u say? people keep saying 27,00. And i never seen one who actualy figured what took me 2 sec, that u should say 7,44 if u think its higher, dont matter if is obvious that is much higher. GOD DAMIT!!! That pisses me off every time i watch! No one ever said just one cent higher or lower!
I still learning poker just to note.

Sean Lind 2009-02-03 18:21:00

No, I don't have any books. But I have more than a books worth of strategy here on this site. Browse around and read it up. If you have any questions or suggestions, just throw them into a comment on an article of mine.


knowledge 2009-02-02 21:28:00

Do you have any books out and if so the names or where can i find them?

Sean Lind 2008-09-12 20:41:00

And I got the answer myself, it was in the way I presented the problem. Thanks for that g3rt, I've updated the article to make that little change :)


Sorry, this room is not available in your country.

Please try the best alternative which is available for your location:

Close and visit page