Mysteries of Bet Sizing

When playing big-bet poker, especially no-limit hold'em in a tournament setting, appropriate bet sizing is a critical variable that solid players use to their advantage, while weaker players use it to beat a faster path to the buffet line. Understanding how to size your bets appropriately is one of the keys to solid big-bet play. And understanding how to interpret your opponents' bet sizes is similarly critical to big-bet success.

No-limit hold'em is the most televised form of poker right now, and it's interesting to see the kinds of bet-sizing errors players make in these games. Of course, it's not always easy to recreate a given player's reasoning. When I see a limit-sized raise (the same size as the previous bet) in Celebrity Poker Showdown, I'm inclined to view it as the kind of egregious error you'd expect from someone whose talents lie in other areas. When I see a similar bet in the late stages of a World Series of Poker event, I'm more often inclined to view it as a sophisticated and carefully considered strategic decision. It's perhaps unfair of me. But it's impossible to divorce interpreting bet sizes from other ways in which you read your opponents.

Although it's important to consider bet sizes in terms of odds, especially in the context of ring games, it's also critical to consider how your opponents will respond to different bet sizes; that is, how they will likely read your bets. This implies that good bet sizing is both context dependent and highly subjective. If your opponents are liable to read a minimal bet as a sign of weakness or as a post oak bluff (a minimal bet intended to create the appearance that you want a call), it might be a clever way to goad them into raising. If your opponents are liable to read the same minimal bet as a clumsy attempt to encourage their call or raise with a dangerous board, it might be a brilliant way to push them out. Even the outcome tells us relatively little. A huge overbet with a strong hand could be the preferable bet, even though your opponent may, after extended mental coin-flipping, decide to pass.

The subjective component to bet sizing makes it difficult to second-guess inappropriate bet sizes convincingly. It's often possible to justify even the strangest bets in terms of reasonable-sounding beliefs about how you thought your opponents would react. It can be difficult to appreciate the factors that go into these decisions, especially if you're not at the table. So, we tend to take a strong player's word for it, while we're more likely to view a weaker player's explanation as a rationalization.

At the same time, many bets are plainly wrong, a fact that's easier to appreciate when you can see players' cards. Most no-limit players know that in a broad array of common situations, making a final call (a call that will close the action) is a much weaker play than making a final bet or raise. Betting or raising all in gives you two ways to win – with or without a showdown – while a final call can win only if you show down the best hand. But in order to make an effective all-in bet, your previous actions during the hand need to be tailored to set it up correctly. Most trivially, if your preflop action doesn't leave you with enough chips to pressure your opponents on the flop, you've given up this strategic weapon. As David Sklansky has pointed out (Card Player, March 29, 2003), it's possible to make this kind of mistake even in limit poker. How many players have we seen on TV bust out because they committed too many of their chips before their opponent saw a flop that helped neither player? When your goal is simply to maximize your expectation on a given hand (as it might be in the early stages of a tournament), disproportionate-sized bets can be an important weapon in building your stack against certain players. This is especially true in the context of fields crowded with inexperienced players, who might be less likely to take proper account of your bet size. If you have a very strong hand and would like to get full value, it's tempting to make the largest bet you think your opponent will call. But in some situations, it may be worth living with a strong chance your opponent will fold. Suppose the pot size is $1,000 and your opponent will definitely call $500. If your opponent is 80 percent likely to call $1,000, your expectation from that bet is $800, a substantial improvement, even though you risk losing the call. If your opponent is only 20 percent likely to call a big overbet of $2,000, the expectation for that move would be $400, suggesting that an overbet is not the way to go. But if your opponent views all big overbets the same way, a larger overbet could be worth it. Perhaps your opponent's bluff detectors will go off to any overbet. Perhaps there's a 20 percent chance he'll decide you're bluffing regardless of how much you bet. If he'll call a $5,000 overbet at that same 20 percent rate, it's certainly worth pushing in more chips, even though you'll come up empty four times out of five.

Online poker adds an additional possibility that complicates the problem of understanding bet sizes. Most of the online sites make it much easier to place a limit-sized bet or raise (the same as the previous bet or raise) than a more reasonable pot-sized, half-pot-sized, or all-in bet. Although the interfaces vary from site to site, it's generally true that to a greater extent than with live poker, online poker bet sizes are influenced strongly by the affordances of the computer programs. With limit poker, bets and raises are in predictable intervals, which means that you never have more than three choices. Bet sizes are also less predictable in big-bet poker because the typical bet size is situation dependent. Pot-sized bets and raises may be common, but players still need the full range of possible bet sizes at any given time.

This adds an interesting wrinkle to the equation. When your opponent makes a limit-sized bet or raise, it could be a post oak bet or bluff, or a weak move from a player who wants to save some chips and buy the pot at a discount. But it could also just be a standard opening bet from a player who has top pair/top kicker, but is too lazy or too preoccupied to manage more than a mouse click. Needless to say, you can sometimes use this confusing situation to your advantage. Other times, it can make your reading problem more complicated, because it adds yet another strong possibility to the mix. With the growing fields in tournament poker, your chances of having close personal knowledge of your opponents are slim. So, you may have to base your read on perilously little information.

Making a small underbet with a vulnerable hand is probably the most common bet-sizing error. Online poker clearly encourages this kind of bet, and the effects probably spill over to live poker, as well. You can benefit from this kind of error when you pick up a cheap opportunity to outdraw your opponents, especially with weak straight draws. At the same time, this makes it much easier to disguise a small bet, intended to build the pot, when you have a stronger hand. When your chances of being outdrawn are negligible, a limit-sized bet is less likely to set off alarms if your opponents might reasonably mistake it for a typical online underbet. If they take a shot with middle pair against your top set, the turn card that gives them trips (and you a full house) can be a valuable card indeed.

Bet sizing is one of the most difficult and subtle issues in big-bet poker, in part because it's really just a subset of the general issues involved in reading your opponents. Although I've generally been an advocate of understanding the arithmetic end of poker, understanding bet sizing is a prerequisite to making good judgments about the odds involved in a given situation.

Daniel Kimberg is the author of "Serious Poker" and maintains a web site for serious poker players at www.seriouspoker.com.