exploitable – Thinking Poker

Complicated Indifference

Andrew — Sun, 04 Mar 2018 14:20:07 +0000

My latest poker strategy article, Complicated Indifference, is now appearing the March 2018 issue of Two Plus Two Magazine. It aims to explain the value of acquiring at least a low level understanding of game theory, even if you know that your opponents will make plenty of exploitable mistakes:

Many poker players are introduced to game theoretical concepts like balance and indifference in the form of a simple example in which one player holds a bluff-catcher and faces a bet from a polarized range. These players tend to understand that there is some sort of theoretically optimal ratio of bluffs to value bets for the betting player, even if they don’t fully appreciate what it is, where it comes from, or why it matters. Likewise, they understand that there exists some optimal calling frequency for the bluff-catching player, though they may not know exactly how to calculate it.

Though not technically wrong, this simple example misrepresents the complexity of the real life situations that arise much more frequently. In fact, most ranges aren’t perfectly polarized, at least not at equilibrium. That is, in the majority of real no-limit hold ‘em situations, there are good game theoretical reasons to bet a variety of types of hands, many of which will not slot into neat categories of “bluff” and “value bet”.

It is in these more complex situations where understanding the game theory behind poker is most valuable. So, while it’s a good starting point to learn the game theory of a simple situation such as the one above, it’s not hard to see why players who understand only this situation fail to see the utility of game theoretical approaches. This article will consider, in broad strokes, the game theory underlying more complex bluffing and bluff-catching scenarios. While it is admittedly not practical (and often not desirable) to memorize and implement perfectly balanced mixed strategies, understanding the basic theory behind these strategies is quite useful for crafting exploitive strategies.

Mailbag: Playing Like a Robot

Andrew — Thu, 12 Dec 2013 16:30:44 +0000

Q: Remember years ago the training websites were saying we want to teach you to not become a “robot” and think about decisions. Wouldnt that be more of the exploitive strategy and now people almost want to play like a “robot” with game theory. Do you find that is happening now and it is interesting how the game evolves.

Another question is that is GTO better or worse for micro stakes players? I think playing exploitive would be a lot better moving up the stakes what’s your opinion on that?

A: Playing robotically and playing like a GTO-approximating poker super-computer aren’t necessarily the same thing. I’ve generally understood (and used) the admonition against “playing like a robot” as a warning against unthinking play, but unless you are a poker super-computer, it would probably require a lot of thought for you to come even close to approximating Game Theoretically Optimal play. Plenty of people, on the other hand, go on auto-pilot in extremely exploitable ways: betting every flop after they raise, folding any time they don’t have a piece of the board, calling just because they have top pair, etc. Whether you’re trying to play exploitively or not, playing well is going to require thinking deeply and situationally.

As for whether you should be trying to play exploitively, that depends on your skill level compared to that of your opponents. By definition, exploitive play requires you to recognize some mistake your opponents are making for you to exploit. When you’re playing with players better than you, that’s tough to do. Weaker players, who are generally but certainly not exclusively found in smaller stakes games, make more mistakes that can potentially be exploited.

A simple example would be a player who hates to fold when he has any piece of the board. If you play micro-stakes, I’m sure you’ve encountered a few of this type! Although you’d make money against this player employing balanced value betting and bluffing ranges, you wouldn’t make as much as you would with exploitive strategy. The most obvious exploitation would involve a lot of value betting, including with a lot of hands too weak to be part of a GTO range.

You’ll find that as you encounter tougher players, which will happen as you move up in stakes, it will be harder to spot such glaring leaks or formulate such obvious exploitive strategies. That doesn’t mean it can’t be done – there are plenty of good and even great players with exploitable tendencies – but they are harder to find and take advantage of. These players will also be more adept at recognizing and exploiting obvious imbalances in your game. You’ll end up playing a more balanced strategy against them as a result.

Do you have a question for the Thinking Poker Mailbag? Please leave it as a comment below!

Surviving a Multi-Day Tournament

Andrew — Wed, 04 Apr 2012 12:53:48 +0000

My latest poker strategy article, Surviving a Multi-Day Tournament, is now appearing in the April 2012 edition of Two Plus Two Magazine. The intention of this article is to get beyond cliches about “valuing your tournament life” to some in-depth advice about how, when, and why survival should factor into your decision-making. I believe that the key is anticipating bad spots and avoiding them from the very beginning rather than waiting until you’re actually facing an all-in bet and then making an excessively tight fold:

The now-tired debate about survival vs. accumulation is nowhere more relevant than in these well-structured events. If you really are one of the best players in the field, then the deep structure will provide plenty of opportunities for you to realize that skill edge. This means that you would do well to avoid close gambles for your tournament life or indeed for a large percentage of your stack, as losing these gambles would jeopardize your ability to take advantage of the many profitable spots you anticipate in the future.

Because this is no-limit hold ’em, the last of your chips could be threatened at any time. Thus, you should be proactive about anticipating and avoiding spots in which you could face a tough decision in a large pot. A good rule of thumb is that if you aren’t going to be comfortable facing a big bet on a future street, then you shouldn’t call a big bet on the current street. If you believe you can profitably get all-in on the current street, then raise; otherwise, fold.

As always, please let me know what you think!

I also want to bring your attention to an article by Philip Newall in this issue that gives a wonderfully clear and thought-provoking explanation of why concepts like “balance” and “game theoretically optimal play” matter and how they can improve your play:

One misconception about GTO is the belief that it’s possible to “solve” a given scenario and come to exactly the right play. This isn’t currently possible; poker is a much too complex game to derive exact solutions for, even with the most powerful computers available. To make GTO practically useful, you just need to imagine your entire range in a given spot, subdivide that range into groups of hands with similar values that’ll be played the same way, and then decide where your current hand slots into that range.

Oftentimes the right play will come to you via a process of elimination. You’ll only take so many actions with a hand in a given part of your range. Taking one action might make your overall strategy either too passive or too aggressive for the strength of that range.

After reading Newall’s article, you might want to have a look at my attempt to discuss similar concepts in the context of NLHE tournaments and dispel some common misunderstandings.

What Are You Afraid Of?

Andrew — Tue, 05 Jul 2011 12:58:05 +0000

My latest poker strategy article, What Are You Afraid Of?, is now appearing in the July 2011 edition of the 2+2 Magazine. The article uses a single hand to look at a larger point about game theory and exploitability:

Being “underrepresented” or “too weak” are not bad things in and of themselves. And before acting on feelings like these, you should articulate exactly how they could end up costing money. In other words, what would your opponent need to do to take advantage of this “mistake” that you are considering?

I’m always curious to know what you think, so please read the full article and leave me a comment!

I’m Off the Wagon

Andrew — Thu, 12 Nov 2009 23:18:28 +0000

My tournaments results have been pretty good the last few months, culminating most recently in two PCA seats and a win in the FTP $300 Saturday 6-Max. I attribute this improvement to a disciplined effort to avoid bringing the aggressive play needed to win in high-stakes cash games over to a tournament setting where my opponents are far more passive and far less tricky.

Last night, in the $300 rebuy FTOPS event, I was rolling along nicely and then fell off the wagon:

Full Tilt No-Limit Hold’em Tournament, 170/340 Blinds 25 Ante (6 handed) – Full-Tilt Converter Tool from FlopTurnRiver.com

BB (t13511)
UTG (t19356)
MP (t20140)
CO (t18855)
Button (t12236)
Hero (SB) (t19919)
Hero’s M: 30.18

Preflop: Hero is SB with 8, K
4 folds, Hero bets t1020, BB calls t680

Flop: (t2190) 9, 2, 7 (2 players)
Hero bets t1333, BB raises to t3125, Hero raises to t18874 (All-In), BB calls t9341 (All-In)

Turn: (t27122) 2 (2 players, 2 all-in)

River: (t27122) A (2 players, 2 all-in)

Total pot: t27122

Results:
Hero had 8, K (one pair, twos).
BB had A, 9 (two pair, Aces and nines).
Outcome: BB won t27122

This is just such a perfect spot for a bluff raise on the flop (I have a wide range, this is a tough flop for either of us to hit, and the stack sizes don’t give me many options even if I don’t believe him). Although I was new to the table and didn’t recognize his screenname, I did have some reason to think he was capable of making plays: he 3-bet me on his button at the first opportunity.

When dealing with an essentially unknown villain, though, shoving any two is not the best way to proceed. I can minimize exploitability by bluff shoving hands that will have the best equity vs. his calling range: draws, hands that paired the board, and even two overcards. Shoving something as weak as K8 makes me easily exploitable by a strategy this guy may well be employing, which is only raising when he has a hand good enough to call a shove. And naturally in a spot where he ought to be raising a huge range he has managed to flop top pair top kicker. AKDHAKLSFHSKLFHASFKH