Playing K-K Against an All-In Raiser

by Daniel Kimberg | Published: Jan 02, 2004

It's an all-too familiar no-limit hold'em tournament scenario. You hold K-K and open for a pot-sized raise. An opponent who has you covered moves all in, and you have to risk busting out to continue with the second-most powerful starting hand in hold'em. K-K seemed like a pretty powerful hand when you raised, but now that you're in the position of pushing the rest of your chips in, it seems like a little less of a sure thing. As often happens in poker, the value of a hand can change dramatically from one moment to the next, often without any cards being dealt.

Many players would simply push all their chips in here, and take their chances, figuring that if they can't get them all in with cowboys, it's not worth playing hold'em. But while K-K is one of the top two starting hands in hold'em, there may be better investments for your hard-earned tournament chips than playing them through an all-in raise. My goal in this column is to address the issue of how much value your K-K has after the raise, and to present a simple approach that may help in deciding what to do with the hand.

Are there really situations in which K-K should be folded? Clearly, against an opponent who would raise all in only with A-A (given whatever else you know), you're probably best served waiting for a better opportunity. Against slightly less selective opponents, it's less obvious what your policy should be. We can get a handle on the economics of this situation by examining your win percentage as a function of what you think your opponent might hold. That reduces your decision to considering how much of an advantage you need to make the call worthwhile (or how much of a disadvantage you're willing to accept). Fortunately, it's easy to calculate the win percentages for all-in bets just by taking the average win percentage of each hand your opponent might hold (including suits, counting ties as half, and remembering that your two kings are out of the deck). The following table shows your win rates as a function of your opponent's raising hands.

We can note a few obvious dividing lines. Most players recognize A-A, K-K, and Q-Q as the three most powerful hands in heads-up hold'em. Unless your opponent would be willing to move in with weaker hands than those, your equity in the pot will be less than 50 percent (almost exactly 50 percent if they always move in with those three hands). 50 percent isn't necessarily the most meaningful dividing line, it's just a convenient point of reference. If you're trying to maximize your expected value (EV) for the hand (say, in the early stages of a tournament, when doing so may be the best way to maximize tournament equity), a better dividing line would be defined by pot odds. If your opponent's all-in bet overbets the pot dramatically, you need close to a 50 percent win percentage to justify calling. But the smaller the all-in bet relative to the existing pot, the better your pot odds for a call. If the pot offers you substantially better than 2-to-1, you may be willing to call even when you know your opponent never moves in with Q-Q, but occasionally likes to play A-K. However, it's worth keeping in mind that if the pot odds are substantially better than even, and your opponent will reliably move in with Q-Q, a call will have positive EV (in tournament chips, or in a ring game).

Conversely, in tournaments there are many situations in which the decision point will be substantially higher than 50 percent. Toward the end of a tournament, especially near a big discontinuity in prize money, playing somewhat more conservatively may be the best money move (albeit one that should be balanced against the prospect of building a bigger stack for the endgame). Early in a well-paced tournament, you may be similarly reluctant to commit your chips without more of an overlay, especially if you expect to have many opportunities against weaker players.

I've worked the chart out for only so many cases. By the time we start including pairs as low as 10-10, the odds strongly favor calling (although you can always manufacture tournament cases in which survival is more important). I've also made life a bit easier by starting with K-K. If you work out a similar table for weaker open-raise hands, you probably need to consider a few more cases.

One striking feature of the win rates in the table is evident in the win rates against players who would move in with A-A, K-K, and A-K. There is a big difference in win percentage depending on whether your opponent would make the move with A-K offsuit or just the suited variety (from 32 percent to 47 percent if you also throw in A-A and K-K). This is slightly misleading, however. A-K suited isn't much stronger than A-K offsuit heads up – either is a big dog to your kings (about 7-to-3). But there are eight ways to make A-K (after we take your cards out of the deck), and only two of them are suited. Players who would jam only with that extra miniscule edge are more dangerous for you here just because there is a much smaller chance they have a hand you dominate. The same would apply for players who jam with A-K 25 percent of the time, regardless of suits.

Finally, if you think something seems hopelessly tight in all this talk about folding K-K preflop, bear in mind that the situations I've described in which K-K really should be folded may be pretty rare. If your opponent's raise isn't much larger than pot-sized, several factors make it unlikely the odds will favor a fold: The pot odds will be great, you probably have your opponent out-chipped, and it's likely a situation in which your opponent will be playing weaker cards anyway. If your opponent has overbet the pot substantially, knowing the odds (and the psychology) should be more important. But it's always important to have some sense, when you consider calling, of roughly what your odds are. It's tempting to talk yourself into playing K-K against any raise just because it's the second-most powerful starting hand in hold'em (before you have any more information) and may not come again for a while. But that big all-in raise potentially conveys a lot of information about what your K-K is up against, and it's worth taking that information into account.