10 Jul 2024 exploit moving up stakes ranges Passive players make lots of mistakes playing poker, and the most recognizable one is their perceived inability to bluff enough. Because of this perceived leak, most people assume that the proper way of exploiting passive players is by over-folding against them. They see a passive player putting money into the pot and just snap fold their hand without much thought. However, this is not the way to maximally exploit passive players. What may seem obvious at first, just fold against them, is only the tiny tip of the iceberg. There's much more you can do to exploit them. After spending years studying with solvers to understand optimal behavior, thousands of hours analyzing population tendencies to understand human leaks, and developing my own automatic node lock-in software to study max exploit strategies, I've developed what I call the "Three Levels of Exploitation Framework." In this material, I'll apply this framework to the context of exploiting passive players and show you how much more money you can make against them. So, let’s go! The First Target — Action Level of Exploitation The first level of exploitation is the level that 99% of poker players operate at. It’s probably the level you operate at right now, and it's the level I mentioned at the beginning of this material. This level is the "Exploiting the Target Action" level. At this level, all you care about is adjusting your poker strategy after the target action from your opponent. The simplest example is when you over-fold against someone you think is under-bluffing. You adjust the part of your strategy that comes after your opponent's action of betting. It is, in fact, true that if your opponent is under-bluffing, the correct response is to start folding many hands you would previously call with. This is particularly relevant on the river, where if your opponent fails to reach the necessary amount of bluffs to make your bluff-catchers indifferent, you should respond by folding all of them. This will happen quite frequently against the more passive regulars in your pool. For example, take a look at the bet-bet-bet line range composition for regulars with a won-at-showdown (WSD) percentage lower than 47%. We can see that for the 75% bet sizing, they have, on average, only 23% of bluffs. When someone bets 75% pot into you, they're laying you pot odds of 30%, which means they need to be bluffing 30% of the time so that your bluff-catchers can break even on a call. Since they're under-bluffing by 7%, all of your bluff-catchers are now negative EV investments. When we input such deviation into a node-locking program, the result is, not surprisingly, that you should fold all hands that can't ever be EV-neutral or chop with them. That being said, be careful with making extreme adjustments based on just your feelings about a particular player or action. Even passive players can end up over-bluffing certain lines due to their poor range construction. One classic example is the check-check-bet line when your opponent is the caller out of position. Let's say you open on the button and get called by the big blind. Your small c-bet gets called on the flop, the turn goes check-check, and then the villain bets the river into you. The same regular profile that we saw under-bluffing the triple-barrel line by 7% actually over-bluffs the 2/3 sizing by 8%. It’s the same player but different lines and different behavior. If you want to exploit passive players, go ahead and do it, but make sure you don't overgeneralize. You should still call those people down sometimes. The second level of exploitation goes a bit beyond what's obvious. It’s less about interpreting exactly what someone did and more about figuring out what that actually means from a broader perspective. In the context of bluffing the river, for example, constructing a balanced betting range requires betting the appropriate amount of bluffs to balance the amount of value bets you have. If you decide to bet for a pot size, for example, and you have 10 value combos, you need five combos of bluffs to be properly balanced. So, you go into the pool of your potential bluffs that you got to that particular scenario with and draw five combos to place into your betting range to make it balanced. The remaining bluffs go into your checking range. What happens when you check? Now, positions are reversed, and it's your opponent who decides on the proper bet sizing to utilize. Based on the amount of value bets he has, he will pick the necessary amount of bluffs to be balanced. You, on the other hand, will have to construct a properly balanced defense strategy against his bet. If you call too much facing a bet, your opponent can exploit you by never bluffing, and then he makes a lot of money with his value bets. If you instead fold too much facing a bet, your opponent can make money by over-bluffing against you, as your over-folding tendency is making his bluffs profitable. There are two ways someone can fold too much facing a bet. The first one is by folding with hands they're actually supposed to call. In theory, imagine you're in the big blind against the small blind, and you get to the river after the flop went check-check and you called a small turn bet. Against another river bet, it's pretty common for the defense strategy to include third and fourth pair type hands and even ace-highs. If you fold these hands that you're supposed to call, you'll likely be over-folding. The second way someone can fold too much is by having an excess of weak hands in their range. In the same situation, imagine now that you're the small blind. You bet the turn, but once you get called, you decide that your opponent is likely to call you too much on the river, so you end up giving up most of your bluffs, which your hands are supposed to bet sometimes. Now, when you check and face a bet, even if you call with the bluff catchers in your range, you will be folding too much because your checking range is too weak. See what I'm talking about? Someone who under-bluffs the river is necessarily weak somewhere else as a consequence. By not betting with hands as they're supposed to, they're weakening their checking ranges, which makes them vulnerable to aggression. You should then exploit their checking ranges with more bluffs. The Second Level — Exploiting the Adjacent Actions An imbalance introduced in a specific strategic option necessarily generates an imbalance in the adjacent actions of that node. If your opponent is betting too thin for value, for example, it makes their checking range unprotected with good strong bluff catchers, which means you can bet thinner for value against their checks. If your opponent slow-plays too much and checks strong hands too often, their betting range becomes excessively weak, and you should over-call and over-raise when facing bets. If you want to start making more money against passive players, you should realize that because they play too passively with their bluffs, their checking ranges are usually complete garbage. This is confirmed by data. When we look at the range composition of passive regulars when they check out of position on the turn as the caller, and compare it to the range composition of bots in the same circumstance. We observe that passive regulars have an excess of 7% of weak hands in their checking range, which causes them to fold 13% more than bots when facing bets. So your adjustment, besides folding when a passive player bets, should be to put them into tough spots by expanding your bluffing range whenever they check. Since their range is excessively weak, your bluffs will enjoy higher fold equity, which obviously increases the profitability of betting. Knowing how weak their range is, they can be over-folding so much that even a pure checking hand can become a pure bet. This is level two exploitation. Level 3 — Exploiting the Previous Nodes We finally get to the final and most important level of exploitation. This level allows you to significantly increase the amount of money you can make with your exploits — I'm talking seven, eight, ten times more money than levels one and two. This level of exploitation is much more advanced, and I would estimate that fewer than one in every 1,000 poker players operate at this level. Despite being advanced, exploiting at this level is not at all complicated. It's just that poker players are not taught to think in this way, but here I am to try and change that. The way to understand the final level of exploitation is by realizing what someone means when they refer to the "game tree" in the context of poker. This term comes from the concept of tree diagrams, a visual representation of the possible outcomes of a problem, very often used in the field of mathematics. In the context of poker, a tree diagram represents all the possible decision nodes of the game and all their respective child nodes and their payoffs: To maximize your EV in poker, you should navigate through this tree by selecting the branches that offer the highest payoffs. If a given node has different children with equal payoffs, this means that the optimal play is to use a mixed strategy. Sometimes we choose option A, and other times we choose option B. In this context, exploiting someone means realizing that a specific children node has a higher payoff than the other children nodes, which therefore makes that one the preferred option and the one that should be picked 100% of the time for maximum payoff. That's what we do when we over-fold against someone on the river. We consider that in that node of facing a bet, the possible options: call, raise, and fold, have different values. Call is considered to be negative since the villain is assumed to be under-bluffing. Raising is out of the question and certainly negative as well. And folding has a payoff of zero. Since zero is higher than any negative number, we choose to fold all the time. When we have the opportunity to exploit someone, essentially what's happening is that a particular decision node has its value increased due to our opponent's mistakes. The magic happens when you realize that the sudden increase in value for a children node has direct consequences for what happens in its upper-level nodes in a tree diagram. The value of a parent node is equal to the sum of all the paths that lead to an end node. If a children node increases in value, then necessarily its parent node increases in value, and all the nodes that came before that in that path. In other words, the value gained from an exploit in a specific decision path is propagated upwards in the tree diagram, increasing the value of the path that leads to that node. This, in consequence, should alter the strategy on earlier nodes for a player seeking maximum payoff. If I didn't make myself clear, let me simplify it for you: the final level of exploitation is the "Exploiting the Previous Nodes" level. At this level, a player anticipates the mistakes their opponents are going to make further down the hand and adjusts their strategy even before the mistakes are made. They then attempt to funnel the hand more often into the node where the actual expected value (EV) can be captured. In the context of exploiting passive players, if you can anticipate that your opponent will play too passively with their bluffs on future nodes of the game tree, this has immediate consequences for your early street strategy. First, your weak hands and bluff catchers will face less aggression, which means unpaired hands will have more opportunities to bluff, and bluff catchers will get to showdown more often. Second, your nutted hands will have less incentive to slow play, as the villain is not putting enough money into the pot with weak hands. To demonstrate this dynamic, I used my automatic node-locking tool to simulate a passive player: I took a specific 7-Deuce rainbow board scenario against a big blind versus small blind scenario and then programmed the passive player to play the second barrel node too passively with 65% more checks than optimal. Everything else in the in-position player's strategy was kept exactly the same, including the optimal turn strategy, so that the solver wouldn't be able to take advantage of any misconstruction in range composition caused by the excess of passivity. The only variable change was how often the out-of-position player faced a turn bet. When I allowed the software to recalculate the maximally exploitative strategy against this mistake, the result was crazy: Almost all hands became profitable floats, and the floating frequency dropped from 33% to 2%. The only hands that couldn't call were the garbage disconnected hands without backdoor flush draws — literally everything else became a moneymaking investment. Another important adjustment observed is that strong hands should no longer be slow-played. Hands like pocket kings, pocket queens, and king-queen went from being mixed strategy to being pure strategy raises. What happens is that now that the out-of-position player is under-bluffing, calling with hands has decreased in value, and raising became the highest payoff option. Final thoughts This concludes the application of the three levels of exploitation framework to passive players. Here's a summary of what we see: In the first level, exploit passive players by folding against them. In the second level, exploit passive players by attacking their excessively weak checking ranges. In the third level, exploit passive players by fast-playing your strong hands on early streets and over-calling with marginal hands on early streets. By applying all these three levels of exploits against opponents, money will start to flow from their pockets into yours. See you next time!
