The positive-sum edge

Why zero-sum thinking is bad for poker--and for the world

Two men the police suspect of being co-conspirators in a massive heist are taken to separate holding cells to be interrogated. The only evidence against them is circumstantial. The police absolutely need a confession to be able to hold them—but they aren’t about to let the men know that. Instead, they do everything in their power to convince the suspects to rat each other out, hoping one or both of them will break. If both stand firm on their innocence, they leave as free men. If one rats but the other doesn’t, the rat gets a reduced sentence and the silent one, a far longer one. If both rat, neither gets a reduced sentence and they go down together. What do they do?

This is the classic Prisoner’s Dilemma, arguably the most famous scenario in game theory. Its classic characteristic is the irony of the payoffs: both prisoners are better off staying silent if they can be sure their partner likewise stays mum, but better off talking if they think there’s a chance their counterpart will talk. Individually, talking ends up being the dominant option—and so, collectively, the prisoners end up far worse off than if they had both just stayed quiet.

The easiest solution to the dilemma is prior coordination—but the whole point of the thing is that there is no prior coordination. So how do you fare in this hypothetical game? Absent external enforcement mechanisms (see: mafia code of omertà as the prime example), the answer is often largely dependent on how you view the world more broadly.

The way out of the PD: positive-sum thinking

The Prisoner’s Dilemma is a mixed incentive game that contains elements of two pure games: zero sum and positive sum. In a zero-sum game, there is only one winner: I win, you lose; you win, I lose. A sports competition is the classic zero-sum scenario. Only one team or player can emerge victorious. In a positive-sum game, all parties can benefit by cooperating with each other. If you want my natural resources and I want your military power, just to create a purely hypothetical example, we would both benefit by working together, a win-win.

A Prisoner’s Dilemma is a choose-your-own-adventure in terms of how you proceed. If you suspect that the world is inherently zero sum and everyone is out to get their own and it’s dog-eat-dog and winner-takes-all and everyone should just look out for Number One, always, you will defect. Defect and lose, no matter what, as your payouts for defection are lower than if you had cooperated: if your partner is also zero-sum, you’re both screwed, and if he somehow protects you, you’re still in for it, relative to being free.

If, however, you believe that people are capable of better than that and that your partner—presumably, someone you trust! otherwise, what were you doing partnering with them to begin with?—is one of those people, you will cooperate. And if your partner believes the same? Bingo. You both emerge ahead. You’re free. All because you believed in the good in humanity, in the fact that society depends on trust, cooperation, and positive-sum interactions. The Prisoner’s Dilemma thus transforms from an insolvable lesser-of-two-evils scenario into a coordination game based on the type of world you think you live in—and want to live in. And the clear winners are the positive-sum players.

Repeatedly, game theory has proven the winningness of a positive-sum mindset. In games ranging from the PD to ultimatum to dictator—the latter two, games where there is no seeming reason to cooperate—people who choose cooperation as the default win over the long-term. In a real-world experiment, when people had a chance to compete with each other in repeat games by building computer programs to execute a cooperate or defect game strategy, the winning program was something known as Tit-for-Tat: start by cooperating, and only defect if the other party does. When the contest ran again, an even better strategy emerged: Tit-for-two-Tats. Only defect if the other party does so twice. That way, you’re protected in case the initial defection was an accident—and you avoid being stuck in a defective spiral. Over and over, cooperation wins.

It wins in the real world, too. Generalized trust is one of the best predictors of how well a society does on factors ranging from GDP to health and happiness. Societies with the highest levels of generalized trust (i.e., positive-sum thinking) do better, by far, and those with the lowest do the worst. The bottom line: when given the choice, positive-sum thinking wins.

Back in 2011, Steven Pinker¹ wrote about his Big Idea of the year: it was on the importance of positive-sum thinking and the fact that it was the key to our progress as a society and a species. He argues that the “enriching and pacifying effects of participation in positive-sum games” far predate formal game theory or the idea of zero- and positive-sum.² And when people realize this dynamic, they emerge ahead. He writes:

People, by neglecting some of the options on the table, may perceive that they are in a zero-sum game when in fact they are in a nonzero-sum game. Moreover, they can change the world to make their interaction nonzero-sum. For these reasons, when people become consciously aware of the game-theoretic structure of their interaction (that is, whether it is positive-, negative-, or zero-sum), they can make choices that bring them valuable outcomes — like safety, harmony, and prosperity — without their having to become more virtuous, noble, or pure.

When people understand that positive-sum thinking can trump zero-sum thinking, everyone wins. They win. Society wins. The world wins. They don’t have to be good people. All they have to do is realize that, in the long-term, what’s best for Number One actually is a positive-sum approach.

A return to zero-sum thinking in the poker world

So why am I writing about this now? Two words: Martin Kabrhel. My poker-playing followers know exactly who that is. To my non-poker-playing readers: congratulations on keeping your mind clear until now, and I’m so sorry to bring this into your life, but it’s important.³ I generally have a policy, as readers of this Substack know, of not saying anything negative about anyone. Positive energy only. But when I see something that I feel is a much bigger deal than any one individual, I have to speak out.

Kabrhel is known for behaviors that some people describe as antics, and others, as intolerable. He takes exceptionally long amounts of time to act on most decisions—a technique known as stalling, that is technically legal but slows the game down and, if deemed excessive, requires the presence of a floor-person to police timing. He talks incessantly, whether he is in a hand or not, doing his best to get under players’ skin to get them off their game. Again, this is not technically illegal—although there actually is a rule on the books against excessive talking. He loves getting into people’s personal space—I’ve had him get up from the table and lean over my shoulder while I was in a hand (and he was not)—and otherwise making them uncomfortable. Bottom line: a basic component of his game play is to use every means necessary to get his opponents off-balance, by being as obnoxious as he can without (usually) crossing the line into prohibited behavior.

While many players, myself included, have spoken out against Martin’s behavior as bad for poker, others have applauded it. The game needs characters, they say! It needs villains! Martin is entertaining to watch! And, if you’re a poker news outlet, the controversy he garners gets you clicks! He’s great for the game because people love the scandal. His behavior may be bad for me, personally, as a player at his table, the argument goes, but it’s great for the game of poker at large because it draws viewers.

Me, enjoying some time at a table that does not have anyone trying to get under my skin. Photo: PokerNews.

That reasoning is profoundly flawed. Let’s take “good for the game” on its literal merits and think through the game theory of the situation. It’s easy to see poker as a purely zero-sum endeavor, the way most sports are. And, indeed, any given hand is zero sum: if I’m winning the pot, you aren’t, and vice versa. In that narrow sense, if I can gain an edge on you by shooting an angle or getting you off your peak performance through my talking or stalling or what have you, I should do it. After all, it increases my chances of winning. Indeed, when I’ve written about angle shooting in the past, people have asked me (genuinely) why I won’t do it if it’s legal. If there’s an edge, why not take it?

Why not, indeed? Because, like a zero-sum view of the world, while that approach may seem like a winning play in the immediate-term, over the long-term it becomes self-defeating—and may actually bring the entire game down with it.

Why poker, like life, is better seen as positive-sum

At a higher level, poker is far better served by a positive-sum framing. The game isn’t a one-shot deal, one hand or one game and you’re done. It’s a repeat interaction that depends on a healthy ecosystem of players for its success. Yes, my ultimate goal is to win. But winning in the long run isn’t the same as exploiting every available edge against my opponents in the short-run. This isn’t a one-and-done competition. It’s a repeat interaction. And in a repeat interaction, reputation matters. Repeat play matters. Players who want to come back to the game because they had a good experience matter.

Imagine being a new player and having your first experience be with someone like Martin Kabrhel. How likely are you to come back? Imagine being a rich recreational player coming to have fun at a high roller—and your first interaction is with someone who is slowing the game to a crawl, getting in your space, and commenting on everything you do. You will take your money elsewhere.

Those who argue that viewership is what grows the game should realize that poker isn’t like most spectator sports, where the audience and the players are separate entities. The game depends on new players who come from the ranks of audience members in order to survive and to grow. What’s good for the players, ultimately, is what’s good for the game. Without new inflows, there is no more poker to be viewed.

From a purely selfish perspective of maximizing long-term EV, behaving in Kabrhelian fashion is the opposite of what you want to be doing—especially as more people think that it’s a good way to play.

Breaking norms only works as a minority strategy

Consider any exploitative strategy, that takes advantage of breaking social norms for success. Like, say, psychopathy. In my book on con artists, The Confidence Game, I explain why psychopathic behavior can be great for successfully getting what you want—a highly effective and winning strategy in most any interaction. If, that is, psychopaths stay at 2-3% of the population, as they always have been. If the number grows, the strategies of psychopathic thinking stop working and become self-destructive. You can only take advantage of other people when the vast majority of them are upholding basic social norms. If that no longer holds, you’re screwed.

In any one game against a single opponent, defection wins against a Tit-for-Tat strategy because in that initial round, where you defect and the other player cooperates, you gain more points. Tit-for-Tat, though, wins in the long run against multiple opponents because the overall payoffs are massively higher than those gained through defection. And Tit-for-Two-Tats emerges even further ahead because you reduce the risk of a defection spiral even more.

As Pinker said, you don’t have to be a good guy to behave in a positive-sum fashion. You just need to realize that simple game theory dictates it.

Short-term thinking will kill poker in the long-term. Amarillo Slim, one of the greatest villains of the game—and what a character—understood this perfectly. “You can shear a sheep a hundred times but you can only skin it once,” he famously observed. Players like Kabrhel are repeatedly skinning the players around them, and everyone who is endorsing and applauding that behavior is complicit, by creating incentives for the exact wrong sort of behavior.

I’m not surprised to see this trend emerge in poker at this particular moment in time. If poker is a microcosm of life, what we’re now seeing in terms of support for the Kabrhelian “antics”—yes, I’m putting the word in quotes, because in my mind, they are not antics; they are signs of antisocial, deliberately destructive behavior—is precisely what we’re seeing with the collapse of social norms all around the country, a return to a zero-sum, us-versus-them mentality that has permeated politics and social relations, both.

It’s a mindset that may seem catchy—yay, America first! yay, me first! screw everyone who is against me!—but it’s petty, short-sighted, and ultimately destructive. As Pinker wrote over a decade ago—and as centuries of evidence have borne out—society thrives on positive-sum mindsets and falls when zero-sum mindsets predominate.

Alas, positive-sum thinking is fragile. It needs to be stoked and tended to survive. Otherwise, we devolve. The rise of the Martin Kabrhels of poker is a symptom of a much greater disease—and the fact that people support him and vilify those who call him out is more evidence of the illness.

Want clicks and short-term gain? Keep promoting the Martins of the world. Want long-term game health and growth? Penalize those behaviors and don’t provide them with a platform.

Martin himself is incredibly smart. He does what he does because for now, for him, it’s a net positive. He’s the only one defecting. If that path becomes unprofitable because the incentive structure changes? He’ll change with it. Let’s hope that happens—both in the world of poker and the world beyond it.

1. Disclaimer: He was my undergrad adviser and is still a mentor and friend.

2. He has a new book coming out in September that delves into some of these ideas more deeply.

3. You’ll note a lack of links here. That is on purpose, to avoid giving a platform to the promotion of behavior I see as deeply problematic.