The past decade has brought a wave of books, articles, and videos about applying Game Theory Optimal (GTO) strategy to poker. There are reams of mathematical calculations to support the potential usefulness of GTO-based poker play. There are also highly skilled professional players able to execute at least some of the strategies to positive effect. In a game of elite GTO savvy players, thought processes and calculations can indeed become very complex, pushing players to the limits of their processing ability.
However, 99% of players at most other levels are not playing anywhere near GTO optimal poker. In the typical poker room, play is often not informed by even a basic understanding of how GTO might be employed. Given these conditions, the application of GTO methods in most situations would likely be destructive to one’s profitability. For example, failing to bet strong hands at a very high rate when heads-up against an unrepentant calling station, just because GTO theory would deem it ill-advised, would be a colossal mistake.
I decided to pick the brain of Steve Blay, co-founder of Advanced Poker Training for some tips on where GTO strategy can go wrong. Steve’s academic background and long career in playing, coaching and programming for APT have produced an unparalleled grasp of the mathematical underpinnings of correct poker play. There is no better person to both appreciate the value of GTO and also understand its gaps.
Steve recounted that in building virtual opponents for APT he “specifically intended NOT to utilize a GTO strategy. Playing against GTO bots is of limited usefulness to most poker players. The opponents on APT are designed to play human-like. They make some of the same mistakes your typical human opponents will make, which is exactly what you want to be training to exploit.”
While GTO theory has some applicability even in lower level games, Steve feels there are many situations where you distinctly do not want to follow the edicts of GTO. Below are four key areas that he identified as times when you should break generally from a strict GTO approach in most games.
- Don’t try to be “unexploitable.” Probably the greatest tenet of GTO poker theory is to be unexploitable. We want our opponents to be powerless in their decision-making, indifferent to whether their actions are profitable. However, this requires that our opponents are playing with GTO in mind. For example, GTO would recommend calling to bluff catch at certain ratios given specific bet sizes. If an opponent bets the size of the pot on the river, game theory would say to call that bet 50% of the time to avoid getting exploited. However, recreational players don’t bluff often enough on the river to yield big bucks with this strategy. That pot-sized bet indicates significant strength far more often than not. In this case, you would not bluff catch on the river nearly as much as game theory would dictate. Yes, by folding you become “exploitable” to players bluffing at an ideal rate. But if there are only 10% pot size bluffs in an opponent’s betting mix, calling 50% of the time is going to get expensive fast.
- Don’t semi-bluff too often on the flop. GTO bots semi-bluff a lot on the flop. Sometimes two to three times as often as they value bet. However, GTO flop semi-bluff theory is scuttled when opponents are calling way too much. And savvy players in every local cardroom in America know that many of their opponents are calling too much. Therefore, you would rather have a higher ratio of value hands than GTO would indicate when you bet the flop. Even weaker value hands like top pair bad kicker can get value from medium pocket pairs and other lesser hands that your opponents will be calling with. In short, against your typical mix of opponents, value hands will get paid more than they should, and semi-bluffs fail to achieve the fold rates they should.
- Don’t “balance your range” pre-flop. A GTO approach advocates moving away from a linear range in pre-flop actions (e.g. only raising with your strongest hands, and calling with moderate strength hands). A classic example is when you open raise pre-flop and then get 3-bet. If you were playing against world-class players, game theory says you would need to 4-bet with some strong hands (AA, KK, QQ) and 4-bet some “semi-bluff” hands such as J-10s and A-5s to balance your play. However, your typical poker room opponents are not 3-betting as frequently as stronger players. Their 3-bet range is generally much stronger than GTO would presume, and therefore you should only 4-bet with premium hands.
- Don’t be afraid to vary your bet sizes. Some pros and GTO advocates say to always c-bet the same amount to avoid giving away information. However, in low stakes games, you should vary your c-bet size based on your goals, because most opponents are inelastic (i.e., exact bet size does not influence their decision). Bet between about 1/2 the pot size and the full pot. C-betting close to the size of the pot with your strongest hands will extract maximum value from players who call with a weaker range than they should. Conversely, semi-bluff closer to 1/2 the pot size to get them to fold out the weakest part of their range at a low cost. Most opponents are too busy texting bad beats stories to their friends to notice the difference.
Different opponents and settings require far different approaches. Throughout his vlog during the 2019 World Series of Poker, Daniel Negreanu spoke of needing to move toward a more exploitative style in lower buy-in, larger field events and to adopt a GTO approach in higher-end events filled by mostly elite pros. Negreanu was relatively unconcerned about being exploitable in $1,500 buy-in WSOP NLH events, but in $100K events he knew he needed a more optimal approach in order to cash.
[bctt tweet=”99% of players at most levels are not playing anywhere near GTO optimal poker.” username=”PokerTraining”]
If a $1,500 buy-in does not necessitate a GTO approach, one can assume that a $1/$2 cash game or $100 tournament buy-in at your local card room will certainly not require it either. Steve notes that “perhaps against a few of the better players, your approach will require the kind of greater balance, deception, and protection that GTO offers. But most of your opponents will lean far too wide on the loose-tight and passive-aggressive spectra. To realize maximum expected value in hands with those players requires exploiting their extreme tendencies.”
Steve and the programming team at APT have developed a range of tools to help improve your exploitative play for maximum profit. In their realistic game simulator, you can create a table against low or mid-level skill opponents in either cash (9-Max or 6-Max) or tournament play that best reflects the typical experience at your favorite card room. Even better, you can alter the playing style of your opponents to focus on specific exploitative styles. Want to focus on aggressive play to steal overly tight player chips? Select options to make the table tighter than average. Conversely, if you want to work on an appropriate range for calling and 3-betting against loose-aggressive players, select the “over-aggressive” option.
Expand the focus of your training even more by selecting a specific table position to practice against these styles as well. How can you take advantage of a passive table from the cutoff as opposed to the big blind? Further, you can layer in specific hand types for even more focused training (e.g. suited connectors in middle position at an overly aggressive table). For tournament practice, you can see how these situations require different approaches depending on whether the tournament is in its early, middle, or late stage. APT’s philosophy is that the only way to master poker is to practice with a specific intent to determine how different approaches are more or less effective.
APT’s reporting gives you key statistics on your play (e.g. VPIP, c-bet %, aggression factor) and explains how those metrics compare to those of average and winning players. APT identifies areas of weakness via weekly Training Plan reports (e.g., being too passive pre-flop, or not c-betting enough). This feedback on your gaps is accompanied by a set of structured tasks through which you can shore up your leaks.
Additionally, APT’s revolutionary Combat Trainer offers various tricky situations for you to navigate from pre-flop through the river. Flop a set, but the board is monochromatic in a suit you don’t have? Run through 100 hands of that situation in about 15 minutes to see how to maximally profit in such situations. APT simulations rotate your table position as well, so you can get a sense of how things play out in position versus out of position.
Steve’s overall view is that “taking full advantage of your opponents’ mistakes is the key to profitability for most poker players. While GTO poker is fascinating conceptually and interesting to watch, at most poker tables your goal is to exploit the weaknesses of your opponents. Making overly tight players fold-out middle pairs when you have nothing, or charging calling stations maximally for that same middle pair when you have an overpair, is going to yield very positive results. “
Bottom line: use all of the tools available to you, at Advanced Poker Training and beyond, to fully evolve your exploitative poker game.
Like this post? Head on over to the sidebar and subscribe. We will let you know whenever a new APT blog post goes live!
