Humans are famously bad at intuiting how often low-probability events occur. People who frequently gamble have intuition marginally better than the general population, but they’re still pretty bad at it. Computers do not have intuition, but they are very good at knowing how often things happen. One of the first solver plays many people learn is three-barreling low pocket pairs on high card boards. If you raise the CO and have 22 on AKT, it’s time to blast off. The reasons to bluff 22 are because 22 doesn’t block your opponent’s check/folds, you can fold out 33-99, and because you have two outs to turn or river a set. After playing millions of hands versus itself, solvers recognize that the ability to turn or river a value bet gives a marginal amount of EV to 22 that makes it a preferred bluffing combo.
Generally when humans hit one- or two-outers, it’s framed in terms of luck. "How lucky am I / is that asshole?” In the grand scheme of things, a one-outer isn’t even that unlikely. Next time you play live poker, pick one of the 52 cards in the deck: I bet it will be the first card you look at least one time on that day. The solver has seen it all; a runner-runner straight flush is just another outcome that drips a tiny amount of EV to one combo in one spot. In today’s hand, I don’t consider the effect of a potential two-outer and cost myself a teaspoon of EV.
Triton Jeju 2025 Event #2 20k NLHE
Level 14: 10k/25k 200k Starting Stack 200k. 85 left 55 cash. 816k Average.
It folds to Tobias Schwecht in the SB who raises to 85k off 1.7M, I call A♥️4♠️ in the BB off 905k.
Flop (195k): 5♠️5♥️4♥️ Tobias bets 90k, I call.
Turn (375k): 4♣️ He checks, I check.
River (375k): 3♥️ He checks, I bet 385k, he calls and mucks.
Keep reading with a 7-day free trial
Subscribe to Punt of the Day to keep reading this post and get 7 days of free access to the full post archives.