selected as the Article of the Month, December 2006
The contest articles can also be viewed at the Internet Texas Holdem Forum

“… for as a man shall sow, so also shall he reap.” (Galatians 6:7)

We normally see the above in a slightly simpler proverbial form, along the lines of: “… as ye sow, so shall ye reap.” However it is worded, the message seems clear: We get what we deserve in the end. There is a clear and definite link between our actions and the consequences. We are in control of our own destinies. Many poker players – particularly those with at least a moderate level of experience and a degree of commitment to the game – are acutely conscious of these sentiments.

Beginning poker players often struggle with the concept that the link between actions and their consequences are often far from straightforward, at least in the short-term. Poker books, articles, and forum posts constantly remind us to remember the sometimes debilitating effect of luck and variance on our short-term results. However, in poker the link between our actions and their consequences is often much more subtle than we sometimes realize.

Let’s take the general issue of bluffing as an example. This illustrates clearly how players at differing levels appreciate the action-reaction link in different ways. Beginning players often overestimate the importance of bluffing. More importantly, they typically view bluffing as a means to an end – as a way of stealing pots to which they are otherwise not entitled. Here, there is a trivial link between the action (the bluff) and the reaction (whether they win the pot). They will only see value in the immediate success – or otherwise – of the bluff. Most importantly of all, their decision-making processes are centered wholly on the immediate hand in question. The decision to run a bluff is based on whether they think it will win them this hand.

As players become more experienced, their view widens a little. They realize that bluffs won’t work every time. Soon, they are comfortable with the idea that a bluff may only succeed, say, one time in three, but acknowledge that one time in three is good enough to turn an overall profit providing they keep making that bluff when faced with the same situation over and over again. This is rather like the beginning-to-intermediate player becoming comfortable with the notion of pot odds. If the right pot odds are on offer, a call can lose most of the time but still prove profitable in the long-term providing the player keeps making the call when faced again with a similar situation.

Incidentally, this type of thinking applied to starting hands gives some players great trouble. A pair of aces on the sort of table where many players are seeing the flop and calling all the way to showdown is extremely profitable in the long term. However, if the table conditions are extreme enough, even AA might win less often than it loses. This means that the hand will prove very profitable overall, but some players believe they are “entitled” to win the pot more often with AA than they do – and they find the lost pots hard to bear. Note that we said lost “pots”. The money is coming the way of the AA player eventually.

But let’s return to bluffing. Players having a yet more sophisticated view than we’ve seen so far, consider the wider value of bluffing. They see bluffs both as a means of stealing pots and of enhancing later non-bluff value bets. They realize that if one is caught bluffing they are more likely to be “looked up” by opponents next time when they’re actually holding a good hand and are betting for value. Here, their perspective has extended not only beyond the hand in question, but also outside the context of bluffing itself. The link between actions and their consequences becomes much less straightforward.

A highly advanced view of bluffing completely eliminates the obvious link between actions and their consequences. In The Theory of Poker, David Sklansky describes a game-theoretical analysis of bluffing. He shows that in certain situations one can deploy bluffs at an optimum frequency to turn a hand that is an underdog into a hand that is a favorite – regardless of any strategy the opponent uses. This is a key difference that sets this view on a higher plane than anything we’ve seen so far; it literally doesn’t matter what the opponent does. The bluffer’s long-term expectation is precisely the same. A key element of this approach is that determining whether or not to bluff is decided at random. If the action is essentially randomized, we’re clearly going to have a tough time figuring out the link between action and consequence. And that’s precisely the point: if it’s tough for us, it’s next to impossible for our opponents.

Let’s summarize this by looking at what we want our opponents to do in response to our bluff. In the beginners’ view, we want our opponent to fold when we bluff. A level up from that, we want them to fold often enough to make the bluff profitable. Next, we would like them to fold but we’re gaining value for next time even if they look us up. Finally, it literally doesn’t matter what they do.

What does all this mean for us in a wider context, as poker players looking to develop our game? It serves as a reminder that poker, despite being simple to play, is a rich and complex game. Of course we should look to fix our leaks, but we should retain a healthy degree of skepticism and be pretty sure of our evidence that something is a weakness in our game before taking steps to correct it. The link between action and reaction in poker is often more subtle than we realize, and this extends far beyond the obvious issues of variance and short-term results.