This week’s mailbag is special because the question was submitted to the podcast, and both Nate and I answered it.
Q: I started out as one of those players you often refer to who says “I don’t really need to balance my ranges in the games I play – my opponents usually suck, and when they don’t I won’t be playing with them long enough for them to notice anything actionable”. After having that position repeatedly pounded on by two people who are smarter than I am about poker, I’ve come around, at least partly, to your way of thinking.
I’m dragging my feet in some spots, though. It seems to me that there are two extremes – against non-thinking/non-observant opponents there’s not much need for balance [I *think* you agree that’s true?], and conversely if I’m McDonald playing against Haxton I just have to suck it up and balance. But in the vast middle-ground I inhabit, especially as I prepare for the WSOP, it seems like I can get away with being unbalanced in some situations as long as I’m aware of what my observant, exploitative opponents have seen from me and can make a decent guess at how they’re likely to respond to unbalanced actions. Obviously the key point there is being “aware of what they’ve seen from me”, so once I’ve played a hand in an unbalanced way but then (successfully or otherwise) counter-exploited my observant opponent, I need to incorporate that into my view of what they’ve seen from me.
Let’s take preflop sizing as an example (I’m not sure this is the best support for my idea, but seems the simplest to talk about). If I come to a WSOP bracelet event table and see Andrew or a tough young Andrew-like hand-reading bastard 2 or 3 seats to my left, I will first silently cringe and wail and curse my misfortune, and then sit down and start to scheme about how to counter-exploit him based on the things I think he will do if I take certain lines. If I open-limp I expect that he will (A) think I’m weak, (B) try to isolate me with a 5x-ish raise with a wide range of hands, (C) expect that I’m likely set-mining with a small pair if I limp-call and stacks are deep enough, and (D) have to worry about me having AA/KK if I limp-reraise on one of the first few times I’ve tried to limp. So let’s say I’m open-limping with small pairs and suited connectors, exactly as Bandrew would expect, but facing an ISO raise from Bandrew I would perhaps limp-reraise my small pairs (repping AA) followed by aggressively C-betting most flops, and limp-call with my small suited connectors and checkraising ragged flops (repping a small set), preferably when I’ve hit at least a small piece so that I’ve hopefully got some equity if called. I’m playing in an unbalanced way – highly exploitable to be sure, but then trying to take advantage of the way my lack of balance will be perceived.
Likewise I’ve heard (just today in fact in the *most* excellent and well worth $19 ThinkingPoker premium podcasts) Nate say that there are lots of players who open 2.5x with their OK hands and 3.5x with their really good hands, so if I see B’nate a couple of seats to my left I’m going to (after the silent wailing) consider doing exactly the opposite (which honestly would be my tendency anyway if I thought I could get away with it) until I think he’s seen enough to catch on, at which point I’ll either move back toward balance, or switch back to what he was originally expecting, or do whatever I think he’s going to mis-perceive.
I can anticipate some reasonable counter-arguments, and I’ve already hinted that preflop sizing might not be the most defensible example. I’m not going to run into exactly Andrew or Nate very often, so I run the risk of making wrong guesses about what Bandrew or B’nate will do to exploit me (but ISO’ing a weak open limper light and correlating a weak player’s raise size to their hand strength are pretty common). I’m also talking about trying to outplay people who are better than I am, likely in bloated pots and possibly out of position, which is generally not a very bright idea, but between the “Kill Phil” value of forcing stronger opponents to play bigger pots than they might want to, and possibly cutting down the SPR drastically so that my tough opponent doesn’t have as many streets to gather info and outplay me, and the value I often get from very good players thinking I’m weaker than I really am … I don’t know, it feels worth considering at least. And of course I don’t have to play completely unbalanced – I can play a sometimes balanced range with occasional diversions into imbalance when I think it’s likely to pay dividends, either from a tough opponent underestimating me or from getting to see a cheap flop with a pair of fours, which *does* have some value no matter how weak the player looks who does it.
Of course another objection could be that my opponents might be so world-class that they just know exactly what I’m doing, and there definitely have been times when an extremely good player has put a soul-read on me that I wouldn’t have thought possible, but generally even then I can attribute it to some lack of attention on my part. If an unknown player did something like what I’m describing, playing in an unbalanced way but either with very different weightings or very different lines than what you would expect, and kept good track of how much information he or she had given out already and how it might be perceived, do you think that you could adjust quickly enough not to get counter-exploited?
A (Nate): One word of caution is that (as you anticipate, probably) I think you misjudge how well experts (maybe not *me*, but experts) will figure out what you’re up to when you try to counter-exploit them. This is something that came up in the episode with Clayton Fletcher–I think he severely underestimated how likely it was that his expert opponents would figure out that he was 3-betting light. Put another way, I think he took it to be too easy to counter-exploit an opponent who he figured would be trying to make exploitive folds against an (perceived-as-)over-tight and scared opponent.
That said, as I imagine Andrew would be quick to point out, figuring out what your opponents are likely to do and then doing the best thing against that set of behaviors is a great way (indeed, the only real way) to think about poker, so hooray for that.
As for “meta-balancing” — there can be times to do something like what you describe, but the central question is really: when and to what extent should we be happy to be unbalanced? I trust that Andrew and I have over and over again answered that question with an emphatic “sometimes and to varying degrees.”
A (Andrew): The key question regarding exploitability vs balance is who will do a better job of guessing at how the other will play. So it’s not just about what have people actually observed you do, it’s what might they expect you to do based on any information that’s available: how you look, what they’ve overheard you say, sometimes even seemingly unrelated plays you’ve made (a lot of pre-flop limping will probably cause your opponents to assume things about your river raising range). So you can get exploited on the first hand you play, simply by people who make good guesses about your strategy – they don’t have to actually observe a pattern to exploit it.
That said, many of your opponents won’t be capable of that. When you believe you can make better guesses you should aim to do so. I consider myself very good at this, but I still find it valuable to understand balance, because it helps me to recognize when people are unbalanced and also helps me to balance in situations where I am NOT sure of the correct exploitive play. That comes up sometimes even against very weak opponents. So it’s not just a matter of “balance against good players but not bad ones”.
All of that said, as I mentioned during the Fletcher strategy segment, there may be times where you are actually better positioned to make guesses about your tough opponents than they are about you. You know exactly who Ike Haxton is and probably even have some idea of what he’ll be doing to exploit some of the weaker players at the table, but he won’t know who you are and will probably underestimate your poker knowledge as a result. You can’t take this to extremes, because part of what makes Ike good is that he won’t let an information imbalance like that persist for long, and he probably won’t do anything too dramatically exploitable on the basis of a weak read (ie “the guy doesn’t LOOK too bright…”), but I do think it’s a potential edge.
If you’re content to play a straightforward strategy of only valuebetting against weak players, you can grind out a small winrate without worrying about balance. But if you (like me) are intent on capitalizing on every edge available, you give away so much, even to weak players, by being unbalanced in frequently occurring spots. Recreational players assume that a 5/6 pot cbet is stronger than a 1/2 pot cbet; they put you on a stronger range when you open to 6BB than when you make it 4BB. But if you’re a tight player, there’s necessarily a smaller sample size of your preflop raises and cbets to analyze, and maybe being maximally exploitative is the way to go.
But at soft low-stakes live tables, waiting for premium preflop hands is leaving money on the table. It’s too profitable to isolate limpers in position, cbet a lot of flops, barrel the turn when their range is weak, bluffcatch when draws miss, etc., and your image nets you even more value when you actually have nutted hands! When you’re playing an active style like this, it’s just too obvious to have bet-sizing discrepancies in respect to hand strength–this is especially true against villains who themselves have bet-sizing tells, as they will project their strategies onto you.
In less common spots, I’m much happier being unbalanced. If you know that villain is never 3bet-bluffing the river, you can check/raise a really high frequency in situations where you assume she frequently bets a wide range. If villain’s range on the river with <1 PSB left is primarily comprised of missed draws that will fold if you bet but bet when checked to (especially applicable to PLO), you can bet all your bluffs and check all of your value hands.
Finally, balance becomes much more important against players with some skill who also have big leaks. To illustrate:
Last night I played a 1/2 session in a raked game in NYC. For the last 4 hours of the night, I had the privilege of being 400-600BB deep against the player to my immediate right. I'd estimate his VPIP/PFR at around 85/15, but he's a pretty good hand reader who's capable of making creative bluffs, calls, and folds postflop. He has very clear betsizing tells (big bets = strong, small bets = thin value and bluffs) and tilt issues.
I isolated his open-limps to exactly $10 at least 15 times and 3bet him (always the same size in comparable situations) half as many times. He never folded or reraised, with the exception of when he was clearly frustrated and limp/raised to $30, at least $10 less than I expect to see him limp/raise for value. I called, raised his flop cbet, and watched him roll his eyes and fold with the disgust a player expresses when she knows she's being outplayed. Against someone like him, splitting my iso- and 3bet-ranges into bet-size tranches would invite him to try to exploit my clearly-defined ranges. Keeping my range wide allows him to put in money OoP with awful visibility.
Wow, that’s a hell of a first comment! Welcome, and great post. (Sorry if it’s not your first and I just didn’t recognize your name)
I’ve listened to 120% of the podcast library, but have mostly lurked the blog (my other post is to thank you for the premium shows). I couldn’t help but post this time because you and Nate have really informed my perspective on this subject.