Learning to Play Poker One Hand at a Time

by Daniel Kimberg | Published: Jan 14, 2005

Some basketball players seem to have the innate ability to score from anywhere, as though there's no place on the court they haven't seen a thousand times. Others are deadly accurate from a few key positions, but are otherwise hopeless, as though scoring is a chore that doesn't come naturally. Both types of players have attributes to admire. But there's something that especially appeals to me about the specialists, the guys who are probably not natural shooters, but who (I imagine) have spent endless hours working on a few key shots to add a dimension to their game. They'll never be Michael Jordans, but with hard work, they develop the weapons they need to help keep the defense honest.

You might think I'm about to launch into an essay on the merits of being a specialist in poker. But I'd like to focus on another aspect of how specialists develop their skills: that long series of practice shots from the same spot on the court. While the Michael Jordans are working the whole court, the specialists steadfastly glue themselves to the small area from which they're liable to get a shot off. They may never see exactly that shot in a game, but the goal is to come close enough often enough to make it a viable weapon.

There's an interesting quality to poker that parallels the differences between these two types of players. Very few of us are natural generalists when it comes to learning how to play poker hands from practice. We hope eventually to figure out which of the 169 starting hands in hold'em are good investments in which situations. But the intermittent feedback in poker makes it difficult to build up a complete picture of how good a decision is. For example, it may not be immediately obvious to novice players that you shouldn't generally play K-9 in middle position for a raise. Learning this from experience is going to be hard. You'll draw K-9 only about once every seven hours of live play (perhaps every four hours online), and perhaps only 10 percent of the times you see it will it be in middle position for a raise. By the time you see it the second time, you're liable to have forgotten what happened the first time. If you decide not to play it, you may or may not get some information about its value. But even if you do play it, you get feedback that's at best loosely related to the hand's true value, and more closely tied to the array of other circumstances involved in the hand, including your opponent's hand, the board, and your post-flop decisions.

Over the course of a few years, you may see the same hand enough times to get a better sense of its worth. But it's hard to connect the dots if you can't really remember all of those details. The best you're liable to get is a potentially unreliable, vague sense that the hand hasn't been good to you. Ideally, it would be great if you could lump all of those hands together in the same session, sort of like the basketball player who spends an afternoon's practice working on that 15-foot jumper. But it's hard to engage in the same kind of concentrated practice that specialists in basketball can. So, poker players have to rely on other tools, like reading and analysis, instead of raw experience.

Of course, you can't afford to be a specialist when it comes to poker hands anyway. You have only limited control over the situations that arise while you're playing, and you can't decide just to muck K-9 all the time because you haven't chosen to specialize in that hand. But you can still benefit from specialist practice. Even generalists have weak spots – shots (in basketball) or hands (in poker) that cause more than their fair share of trouble. In hold'em, you're dealt 7-2 and K-10 equally often. But you really don't need much practice with 7-2 outside the blinds, while trying out different approaches with K-10 could have some benefit. For the same reason NBA sharpshooters don't stay late after practice working on their layups, you can save a bit of time by working on the hands you suspect are costing you the most in bad investments and missed opportunities.

So, how do you get concentrated practice on a single hand in poker? Last year I suggested that you, as part of your poker spring cleaning, pick a particular hand and try to learn as much as possible about how it plays in a variety of situations, about its percentages in different situations, and about how different poker writers have suggested playing it. Consider using a program like Turbo Texas Hold'em to try out different scenarios or just give yourself lots of repetitions. And if you play online, you have a potentially rich database of information about how each starting hand has treated you.

Although I wasn't thinking about basketball when I suggested becoming an expert in a single starting hand last spring, the analogy seems apt. Even if you're a pretty good generalist, it's still worth picking out the single hand that gives you the most trouble and finding out what makes it tick. The worst thing that can happen is that you won't discover anything to adjust your play. But I'm willing to bet you'll be able to draw on your body of explicit knowledge the next time you face a difficult decision involving that hand. And if you're so inclined, take the time to work on your baseline jumper, as well.

Now that I've made this suggestion twice, it's probably time for me to set a good example and do my own hand dissection. I'll put it off a little longer, but I promise I'll take a stab at a detailed starting-hand analysis for an upcoming column – perhaps in time for spring. All I need is some idea of what hand to do. So, feel free to e-mail me requests; I'm leaning toward Q-J in hold'em, but there are lots from which to choose – 169 in total, although most won't come up often outside the blinds.