POTD #238 Winning Wednesday: AA vs KK
The ultimate cooler in a 2 million dollar (HKD) tournament.
Four-bet pots are some of the trickiest poker hands to play and can lead to large EV losses when playing vs. the solver. The strategies are tricky and specific, and the pot is so large that any little mistake could cost you several BBs in EV. Tournament players are especially bad at playing them, because as the tournament progresses, there is a lot less four-betting non-all-in as stacks get shorter, and a lot less flatting of four-bets as ICM pressure discourages people from peeling a four-bet for a quarter of their stack.
In POTD #190, I wrote about a bizarre cold five-bet pot between Jason Koon and David Coleman, who both had premium hands at the Poker Masters Finale Final Table. In today’s hand, I will write about a much more conventional four-bet pot I played vs. James Chen in the Triton Jeju Main Event in 2018. In both hands, a lot of money went in preflop, the flop was QTx, and we needed to navigate how to play a tricky narrow-range spot.
In addition to pairing today’s POTD with a previous hand played by other people, this is a Winning Wednesday, and while I’d eventually bluff of his bullet to David Peters in POTD #228, I did much better on my second bullet and ended up final tabling this tournament, which was at the time the highest-stakes tournament I’d ever played. If the Winning Wednesday and pairing with POTD #190 hat on a hat is not enough for you, I’ve put another hat on the hat on a hat: This is part of a new POTD feature where I revisit old hands I played and, before I run them through a solver, I write about what I think about the hand now. Seven and a half years later, I’ll be able to determine if I punted at the time and if I learned anything from the hand.
Triton Jeju 2018 - NLHE Main Event 2M HKD $255K
(1k/1.5k/1.5k) (SB/BB/BBA) Registration is open. Starting Stack 250k.
It folds to me (263k) in the CO with A♣️A♠️ and I make it 4k, it folds to James Chen (232k) in the SB who makes it 16k, I make it 45k, he calls.
Flop (93k) Q♠️Q♣️T♥️: James checks, I bet 15k, James calls.
Turn (123k) 5♦️: James checks, I bet 40k, James calls.
River (203k) K♦️: James checks with 132k back, I check back and lose to K♥️K♣️.
What I Was Thinking
Preflop, you often trap AA in position, but not when you’re 150 BB deep, and certainly not vs. James Chen, who is live to make some very aggressive, off the grid plays. On the flop: I was not thrilled about this flop, but I figured I still had a range advantage and AA was often the best hand, so I was going to make a small c-bet size that would allow me to bet the turn and shove the river. 15k seemed reasonable — this might be a tell on 2018 me, but I was not going to mess around with small chips in a four-bet pot and bet 13.5k or 16.5k. On the turn, I still felt I had the best hand; I wanted to keep putting money in the pot with the best hand and wanted to set up a comfortable river jam. On the river, I felt there were two problems. One was that James occasionally had KK, AJ, or J9, which made it harder to get called by worse when I shoved. The other was that AK was a common bluff for me, and the K river made it much harder for me to bluff and much less likely that James would call me down with a hand like AT or JJ. AA’s equity decreasing and having fewer bluffs to pull from meant I felt I needed to check back my weakest value bet, so I checked and saw that I had gotten two-outed and my river check saved me from losing an even bigger pot.
What I Think Now
No notes on preflop. I need to four-bet and my size is the correct one. I feel that my flop size is likely too small. I’ve since learned that a common mechanic in three- and four-bet pots is, when you have hands like AA, you want to bet a larger, more polar size, such that you make hands like 88, AJ, or Tx close to indifferent to calling a bet, and so you charge them the maximum instead of letting them see a turn and river cheaply so they can see if they’ve improved before calling a big river bet. I’d rather have bet something like 35k on the flop. However, once I sized small on the flop, I like my turn size, more or less. Maybe 50k-55k to set up a half-pot-sized river shove would be better. On the river, my guess is, the solver is not crazy over checking here because James should always 5-bet KK preflop; however, I bet the actual solution will feature a whole host of odd bluffs I’ll rarely have… like occasionally bluffing 65s. So I think I was likely playing too tight on the early streets of this hand, and checking the river a little too tight was a fine adjustment given the situation.
What I Got Wrong
I could have trapped AA preflop, but it mostly four-bets, and my size is fine. 2026 Sam is right about the flop; my size is too small. I mix some checks, and the solver picks 30k over 15k, but the EV loss is negligible. Despite my flop size not losing EV, the solver really does not like it, because when I force it to bet 15k, it picks a bigger size on the turn to correct for betting too small a size on the flop, betting 60k to set up around half pot to play on the river. It seems like no matter what sizing options I give, it wants to set up half pot to shove on the river. On the river the general shape is, I shove if I block combos of AQ suited and I check if I don’t. I have the one combo that doesn’t block a single combo of AQs, and I ran into KK, a hand that is supposed to pure five-bet shove preflop. James is a loose player, and generally I am more inclined to value bet thinner vs. loose, splashy players capable of hero plays, but there are occasions where looseness and unpredictability tend to give your opponent many better hands, but not many weaker hands, and I think this might be a case like this. A sticky player like James might have more suited Qx and more AJ, and calling down with Tx or JJ here is rather ambitious, especially on the king river.
I think my flop and turn size are too small and I left myself with too much to play on the river, but I think my river check is the correct adjustment. Shoving any AA combo is thin in theory, and if you give him a couple extra combos of AQo or KK, that’s enough to make my river check best.
Types of Error
Sizing error: Did not put enough money in the pot with AA in a four bet pot
Grade
I think it’s notable that while my flop size loses almost no EV, my turn size loses around 750 chips depending on my exact combo, which is not a ton in a four-bet pot, but is not nothing either. My flop size doesn’t lose EV because the solver can correct on the turn; my turn size loses EV because my “correction” on the river is going all-in for more than I want to with a hand like AA. So my turn size is bad, and because I was unable to play the right turn size, my flop size is also bad. When there are 62bbs in the pot preflop, every extra chip you can put in the pot while your hand is best is worth doing. It makes sense that I want to make hands like JJ and Tx put in as much money as possible on the flop and turn, not let them see a river for cheap and then fold a pure bluff-catcher.
My 2026 self overstated how many odd solver bluffs I have on the river with hands like 65s or 98s; it’s mostly just A2s and A3s, because on the king river my suited Kx bluffs river a pair. So overall, I think I failed to put enough money in the pot before the river, and even vs. a player known for his heroic plays on all ends of the spectrum, I think I like my river check, but it’s very close.
I’d say I have two clear mistakes in this hand, my flop and turn sizes, but they were sort of the same mistake, because I could have corrected my flop mistake on the turn but did not. A tough grader would say I should be docked even more for failing to correct my mistake when I had the opportunity to. A lenient grader would bucket them together as one mistake. As I’m writing this, Canada just lost the gold medal hockey game, so I’m going to be nice to myself.
B-