When can you donk bet?

2024-09-18

Today, we're going to look at donk betting in heads up no limit poker. To help contextualize this concept, we'll look at an example hand which was played between a human and a game theory optimal bot. This hand was played on ActionFlop, which lets you play poker against game theory optimal bots while keeping track of your mistakes (click the button above to try it out — it's free).

Before we dive into the action, let's quickly define what a donk bet is. A donk bet is when you were not the preflop aggressor, and you lead out with a postflop bet from out of position. Now let's dive into the preflop action.

The preflop action looks pretty standard. The bot raises from the button and you call with K2o. K2o is a hand which should call a raise 100% of the time.

Let's briefly consider the ranges before cruising ahead to the flop. The BN raises with over half of his range. Many hands mix between calling and raising, but the raising part of the range is stronger. Thus, the BN's range going into the flop is wide and uncapped. It includes all of the pocket pairs. Almost every hand other than a handful of the lowest offsuit holding are there in some proportion.

Now let's look at the BB's range. The BB's range is wide and capped. Because he would have raised with his pocket pairs most of the time, the only pocket pairs which occasionally show up in his range are 88 or worse. Offsuit combinations are overrepresented compared to suited combinations.

Let's keep these ranges in mind as we proceed to the flop.

The dealer crisply unveils 5c9cKh. As the BB, are you thinking about donk betting? In case you aren't familiar with the general rule, it's that you should almost never donk bet. On this flop, your range should check 96.4% of the time. It's not even worth trying to develop a donk betting range on this flop. What are you going to do, roll dice and bet if you get snake eyes?

The rule applies to the vast majority of flops. But you should know that there are some rare exceptions. Think about the ranges we talked about going into the flop. What kind of flop would be relatively beneficial to the BB's range? If you said lower rainbow flops, it's a good first guess. On a flop like 842 rainbow, the BB should donk bet 9.3% of the time, mostly with small bets. That's something, but it would still be fair to simply always check.

By the way, on the flip side, there are flops where you can't even pretend to entertain donk betting. Take AQ7 monotone, for example. You should check 100% of the time. If you forgot your cards, don't even bother looking, just check.

Let's get back to the original question. Are there any flops where you can actually entertain a donk bet? The answer is paired boards, and in particular low paired boards. For example, Kc3c3d is a board your range should donk bet 32.3% of the time. That is meaningful. The reason this board is so attractive to donk bet is the way it interacts with your range. Whereas on the Kh9c5c flop, your EV is about 35% of the pot, on the Kc3c3d flop, your EV is about 44% of the pot. Your range hits trips and full houses on this flop, and more frequently than the BN's range does.

By the way, the higher the board pair, the less often you should donk bet. On 7c6c6h, you should only donk bet 15.7% of the time. On Qc7c7h, you should only donk bet 14.2% of the time. That's getting into "maybe you should always check" territory.

Back to the Kc3c3h example, how should you devise your donk betting range? As you can see, you spread your bets across your entire range.

If you look at the strategy across your distribution of hand types, you can see that when you have a Three, you should bet most of the time. Otherwise, you should bet less frequently. If you have nothing other than the pair on the board, you should check 83% of the time.

Let's finally return to the original hand. Do you donk bet?

You decide to violate the general rule and donk bet. In this particular case, it only costs you about 0.1 in EV, which is about 5 BB per 100 hands. Let's go ahead and continue analyzing the hand. Here comes the turn action.

On the turn, you decide to bet again, and the BN calls. Since you took an unorthodox approach with the flop, let's talk a bit about how you should be thinking about the turn depending on the line taken on the flop. The mainline for the flop action is check-bet-call. The other common line is check-check. In the mainline scenario (flop: check-bet-call), you should still check the vast majority of the time on almost any turn card. Can you guess what the exceptions are? Think back to the discussion about what flops you could entertain donk betting. Nines and Fives, which pair the board, are the exceptions.

What about in the check-check line? In the check-check line, the BN weakened his range by checking back. In this case, you get to wake up and do some betting on any turn card. On this particular turn card, you should bet with 35.8% of your range, with a healthy mix of small and large bets.

Finally, let's consider the flop donk bet scenario. Your tiny donk betting range is fairly strong, with many Kings and various draws in the mix. As a result, you actually get to continue with a turn bet 46.3% of the time, and the vast majority of these turn bets should be large. As it turns out, with your KXo range, you mostly bet large, with the exceptions of K4o, K3o, and K2o. So you do make the correct decision by betting small, and we proceed to the river.

On the river, you induce a fold, which is great, but let's take a closer look at your decision, because it was actually a mistake to bet large. Consider your strategy for your range distribution.

Could you have made the correct decision and bet small here? You have some very strong hands in your range (flushes and straights), some strong hands (sets and two pairs), some value hands (pairs), and some nothing hands. This is a great spot to have a betting strategy with small and large bets. You want to extract value from some of your strongest hands (straights and sets), and then you get to mix them in with bluffs from the bottom of your range. That leaves you with a value range that bets small, mixed in with some more bluffs from the bottom of your range. You can't just bet big with only your strong hands (plus bluffs) and bet small with your value range (plus bluffs), otherwise the BN can raise you off your value range too easily. So you also include the strongest part of your range (flushes) in your small-bet range. Where did your pairs of Kings end up in that? In the value part of the range, so they should bet small.

Let's take a look at the final result.

The BN had a pair of Fives (which he folded). Interestingly, with this particular hand, he should fold to your large bet 100% of the time, but he should fold to a small bet only 48% of the time (mixing between calling and raising the rest of the time). So you get a little lucky with your big bet, but strategically it was not the game theory optimal choice.

That's all for now. I hope you enjoyed this hand analysis. In case you haven't already joined, don't forget to join the newsletter below. Newsletter subscribers receive exclusive poker strategy content which is not posted to the blog, so it's important to subscribe if you don't want to miss the juciest analysis.