Sizing Up the Opposition

Sizing up the opposition is a fundamental poker skill. If you want to make good decisions, you have to watch your opponents play and try to figure out what they are doing. The problem is that so few hands go to showdown, so you rarely get the opportunity to see your opponents’ cards. Even more frustrating, when you do see cards, you often won’t learn much, since most hands shown down are either big hands that were bet all the way or weak hands that were checked all the way. Most people play big hands hard and small hands soft, so the information isn’t that helpful.

You have to derive information from the hands that didn’t get shown down. You do this by observing the frequency with which someone takes a certain action. If you see a player play 20 hands in three orbits, you know he’s loose. If you see a player play two hands in three orbits, you know she’s tight.

Whenever we draw conclusions about how a player plays based on observations, we are unconsciously using a process called statistical inference. We get bits of data here and there, and from the data we draw a conclusion about the process that produces the data. For instance, by seeing how your opponent has played a certain number of hands, we draw conclusions about that player’s overall strategy and how he might behave in a future situation.

Here’s the problem: Statistical inference is tricky, and human brains aren’t wired to do it correctly. Specifically, we tend to give too much weight to recent occurrances, and we often don’t correctly take the characteristics of the population into account.

Here’s an example of statistical inference in action. Say you’re walking down a busy street and you see a man pass you who looks to be about 6 feet 4 inches tall. You might think, “Wow, that’s a tall guy.” Next you pass a man who looks to be about 8 feet 4 inches tall. This time you would likely think, “He must be wearing stilts.”

Why the difference? Both are legitimate observations — this man looks 6 feet 4 inches, that man looks 8 feet 4 inches — but the inference is different. One guy is tall, while the other one is cheating.

The answer is in the distribution of the underlying population. Men who are 6 feet 4 inches tall are uncommon — perhaps only a percent of the total population — but they are not so uncommon that you wouldn’t expect to pass one in the street every so often. Men over 8 feet tall, however, are so rare that there have been only a handful in all of recorded human history. The odds that you would run into one by chance on the street are vanishingly small.

Let’s get back to poker. A friend of mine recently related a hand he played at $1-$2 no-limit hold’em in a locals card room in Las Vegas. He wasn’t sure about what to do on the turn, and according to him, the problem was that he was in the hand with an “extremely good player.” According to my friend, his foe had been outplaying him all day. The guy had superhuman hand-reading skills and was aggressive at all the right times. My friend wanted to know how to adjust to an opponent like that.

I immediately cried foul; he may as well have told me that he had been in a $1-$2 game with an 8-foot Martian.

“You’re wrong,” I said.

He gave me a puzzled look. “About what?”

“About your opponent,” I said. “He didn’t have super-human hand-reading skills, and he wasn’t outplaying you. Based on what you’ve told me so far, I wouldn’t make any adjustments at all.”

It’s the population of $1-$2 no-limit hold’em players. The vast majority of $1-$2 no-limit players are not aggressive enough. Some are too loose, some are too tight, but nearly all of them don’t bet or raise enough.

A small minority of $1-$2 no-limit players are wildly overaggressive. They bet and raise at nearly every opportunity that seems even remotely reasonable. They also play lots of hands.

The number of people playing $1-$2 no-limit hold’em at a locals casino in Las Vegas at any given time who read hands and run bluffs like Tom Dwan is zero. It was virtually impossible for my friend to have sized up his opponent correctly.

My friend said his opponent was playing tight. I told him that meant his opponent was likely a bit on the nitty side, and certainly wasn’t anyone’s world beater. It was just my friend’s bad luck that this guy happened to pick up big hands every time they played a pot together. You don’t need dynamite hand-reading skills when you keep getting dealt the nuts.

My friend’s opponent wasn’t 8 feet tall, he was wearing stilts. At a poker table, catching hand after hand for a few hours can make you look a whole lot bigger than you actually are.

What’s the point? When you’re sizing up your opponents, be much quicker to assign them common traits than rare ones. See a woman play four hands in a row? OK, she’s loose. There are plenty of loose players at $1-$2, and the observed data suggest this woman is among them. See someone slowplay a set on the flop? OK, the player is a bit trappy. Again, trappy players are a dime a dozen at $1-$2.

However, if you see a player shove all in on the turn three times and win the pot each time, do not assume he likes to bluff-raise the turn. Very few $1-$2 players like to bluff-raise the turn, and therefore the observed data do not yet support this conclusion. With only three data points, the far more likely explanation is still that the player is an average $1-$2 player who caught three big hands. With nine opponents, you can expect someone will run hot during your session.

Sure, it’s possible that Tom Dwan’s secret protege sat two to your left in your $1-$2 game during double points happy hour just to torture you, but it’s really very unlikely. Put your observations in context, and above all, don’t let any opponent get into your head. ♠

Ed has authored six poker books and sold more than a quarter million copies. Ed’s newest book, Reading Hands at No-Limit Hold’em, will soon be available for purchase at notedpokerauthority.com. Find him on Facebook at facebook.com/edmillerauthor.