Human Nature vs. Poker Strategy

There was something I was reading about the other day that made me think a little bit about how human nature sometimes goes contrary to correct tournament strategy. It had to do with something in economics called “the reflection effect”. Basically, it’s spelled out like this…

The Reflection Effect


Suppose you have a grand… one thousand dollars. Someone comes up to you with a gift. He says, “Hey buddy… here’s a present for you. You get to make a choice. Either I’m going to give you $500 or we’ll flip a coin and if you win you get $1000.” Expected value wise, it should be the same whichever you choose. Your EV is +500. But the studies show that people are way more likely to take the sure gain. In economic terms, people are risk-averse when it comes to positive gain.

On the other hand, suppose you have a grand… one thousand Washingtons. This time the guy comes up to you and says, “Hey buddy… bad news. You’re going to have to pay me something. I’m going to either take $500 OR you can flip a coin and if you lose you pay me $1000… but if you win, you get away scott-free and don’t have to pay me anything.” In this case, the expected-value is again the same. But it’s human nature to take the coin-flip… that is, people are risk-seeking when it comes to negative loss.

People are generally risk-seeking when it comes to positive gain and risk-averse when it comes to negative loss.


So what does this have to do with poker? Good question… and I’m still wrapping my head around it. But basically, I think it has an effect in poker tournaments. And basically, what I’ve taken from the idea is that people in poker tournaments are more likely to gamble away all of their chips than they should be and less likely to bet their hands for full value when they have a sure winner.

Let’s take it a little further. It’s fairly accepted these days that in a poker tournament, the expected value of your individual chips decreases the more you have. Put another way, if you started a thousand dollar buy-in tournament with a thousand in tournament chips (so each tournament chip had exactly a real-world EV of $1), if you then doubled up the first hand your EV wouldn’t exactly double… it would be a little less than $2k. The reason for this has to do with the payout structure, since if you win a tournament and get 100% of the chips, you usually get more like 20% of the prizepool. This also mirrors real life… any gambler will tell you there is much less of a difference between $10,000 and $5,000 than between $5,000 and $0.

In a poker tournament, the expected value of your individual chip units decreases the bigger your stack gets.


Now suppose you have pocket aces and the flop comes A95 and you’re deciding what to bet. Somehow you KNOW that your opponent will ALWAYS call if you bet half the pot and that if you bet the full pot, he’ll pull out a coin and flip it and call 50% of the time. The reflection effect would say that it’s human nature to go with the half-pot bet. And this is actually correct tournament strategy as well, since the first half-pot sized bet is worth more in EV terms than the full pot sized bet.

But where we see human nature going AGAINST proper tournament strategy is when you’re faced with having to lose your whole stack… those situations where you raise and someone then pushes over the top of you… they have you covered but you’re getting the right odds to call even though you’re not a huge favorite. The point is (I think) that the reflection effect would say that it’s human nature to be risk-seeking in those types of exchanges. And that goes against proper tournament theory. We should be more risk-averse to losing chips than we probably are.

We should be more risk-averse to losing chips than we probably are.


I guess I could have summed that all up with something I heard from Mike Mizrachi once at a poker tournament table in Tunica (the one he ended up going on to win back in 2006)… always leave yourself with something to grind.

This is my site. If you want to start a flame war, go to 2+2.

No Comments

Leave a Reply

Featured