I’ve been reading Philip Newall’s book The Intelligent Poker Player (published by TwoPlusTwo). His book is about game theory optimal strategy, and is especially applicable to short-handed and heads-up limit Hold’em. I plan on writing a review of it soon, but in the meantime I’ll just say that I recommend it. Not only does he talk about strategy, he also has really good advice on playing poker for a living and on the state and future state of online games as he sees it.
I follow Newall on Twitter (@pnewall) and I got the idea to ask him for a critique of my book Reading Poker Tells. It was a little late (I’d already sent the book files to the printers a few days before), but I am always open to criticism and feedback,. I can always update later editions or make changes for the e-book version. Newall agreed to take a look on it, and his feedback was so good I wished I had sent it to him a long time ago.
Newall’s most important feedback was that I should have differentiated more between the application of tells at limit and no-limit. He said: “You could mention that tells in limit might need to be more accurate than in no-limit. If you’re getting laid bigger odds by the pot then the accuracy of your tell has to be higher to sway your decision.”
He is exactly right, and it was such an important point that I should have concentrated on it a lot more. I do have a few paragraphs about the differences between no-limit and limit in this regard, but not near enough. It’s one of those things where something is so obvious to you that it completely escapes your mind to delve into it much. So I wanted to devote this blog post to elaborating on this idea.
The last thing I want readers of my book to think is that they should be making a lot of decisions in limit based on reads. The truth is that it is very seldom that I base a poker decision on a read in limit (it’s more common in no-limit but still not very frequently). For instance, I’ve played 10+ hour sessions of $15-30 for the past 4 days, and I’d say there were maybe an average of 5 hands per session where my decisions were swayed by a read. I see a lot more tells than that; in fact I see several meaningful tells every round in a typical game. But because I’m only in a few pots to begin with, and of those pots I’m in, there’s only going to be a few where a tell is so reliable that I’m going to base a decision on it. Most of the time, my decisions are based on fundamental strategy. Occasionally I will see a behavioral pattern that I’ve correlated to strength or weakness, or else I’ll just notice someone lazily giving away info (like telegraphing their intention to fold or call), and it’ll influence my decision.
If you are making a read-based decision, it will usually be a call or fold decision on the flop or river; one of those borderline decisions that’s probably 50/50 and you just need a little something one way or another to push you in some direction. For example, if you’re on the river with a medium-strength hand, and you check and a tight, predictable player bets into you, you may instinctually feel that calling this bet is a break-even proposition. And if you’re experienced at limit, then the underlying math would probably support it. Let’s say the pot is offering you 9:1 and you think this player is going to have a better hand than you more than 90% of the time. That’s a break-even proposition. So if you have even a slight bit of physical read to go on, even if the read is only 60% accurate, this will make your decisions in this spot better.
For example, let’s say this tight-predictable player has a fairly reliable tell that when he’s value-betting, he tends to stare at his opponent. (That’s a fairly common tell for many people; whereas those people would avoid eye contact if they were bluffing.) If that’s a behavior you’ve correlated with strength in the past, and you see it in this fundamentally break-even spot, and it’s at all reliable, then you have more information to make the fold. Of course, the read may be wrong. Most reads are not 100% (although I’ve seen some emotion-based ones at no-limit that are near 100%); in limit I’d say I’m lucky if I can spot reads in the 80% reliable area. But the point is that a little bit of information goes a long way in break-even spots.
Newall is completely right that because the pot is usually offering you such good odds that you have to be quite sure of your read to act on that read. If you didn’t feel the spot was a break-even one, you would be less likely to rely on a read; for example, you’ve got a medium-strength hand but you know it’s quite likely your opponent is capable of a bluff on the river. Because the pot is offering you such good odds, and your opponent is bluffing a good amount of the time, you’d have to have a very, very reliable read to make that fold. That honestly does not happen frequently.
It sometimes happens that I’m in situations where I think I might have a read, but if I’m not very sure of the read I can’t act on it, because making the wrong fold in limit is catastrophic, because the pots are so big in proportion to the bets. This is a crucial point in understanding how to work tells into a limit strategy, and I regret not focusing more on it in the book. I’m going to write about a couple limit Hold’em hands where reads came into play in the next blog post.
In summing up, thanks again to Philip Newall for bringing that shortcoming to my attention. And I recommend following him on Twitter. He’s been very receptive to questions I’ve sent him about his book.