Know When to Fold ’em: Relationship Advice from the World of Poker

The other day I told a friend who is struggling with a relationship problem about a poker game logic puzzle I used to invoke back when I was treating psychotherapy patients. He found it quite helpful so I am passing it along for whatever it might be worth to others in similar situations.

The logic problem involves a table at which four people are playing poker. You walk around the table and espy everyone’s hand. The first player has a royal flush, the second player has a full house, the third player has a pair of 7s and the fourth player has a king high.

Question: “Which player has the worst hand?” (Answer after the jump)

Answer: The second player, who has the full house. The third and fourth players are better off because they can easily perceive that their hands are losers. They will fold immediately without losing much money. But the second player has an ultimately losing hand that looks like a big winner and will therefore go all in and be wiped out.

Some of the hardest relationship dilemmas have are like the second hand. She loves her fiance in every way and he loves her, but she absolutely wants to have children and he had a vasectomy because he absolutely does not. Or your girlfriend/boyfriend is wonderful and fabulous and loving with you 5 or 6 days a week, but on the other days is horribly abusive. In these sorts of relationships, it hurts like hell to “fold” because so much is good about it and because you wish against reason that it can become a clear winner. But in hanging on you are just setting yourself up for an even more horrible loss down the road.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

19 thoughts on “Know When to Fold ’em: Relationship Advice from the World of Poker”

  1. Seems to me the answer is contextual. Knowing the hands of all players, this is true. The Full House is the worst hand, assuming seriously incomplete knowledge. And we almost always have incomplete knowledge, and one of the most important parts of this lack of completion is knowing how serious the consequences arising from what we do not know.

    Knowing only the hand you have, the full house has good reason to stay in the game even though the ultimate result is disastrous. It much more likely would have been fabulously successful.

    Consequently I think the extrapolation of this problem to relationships does not work.

    1. I might have the full house in my relationship, and someone else has the royal flush in theirs. How does that reflect on my relationship? Should I fold on my not-quite-optimal situation because someone has a “more perfect union,”? I don’t think so. This exercise has more to do with your definition of the words “best” & “worst,” than it does with poker or relationships.

    2. I think you are being too harsh, and somewhat obtuse.
      The larger point is that when you have a good hand, it is easy to imagine the wonderful things that could result from your hand, things that are not guaranteed — and that in fact will NOT occur given what we know.

      So, in the relationship case we are NOT talking about “she wants kids and he has kept his vasectomy secret”.
      We are talking about “she wants kids, he doesn’t want kids and can’t have them, and both know this”. You are now setting yourself up for a dynamic where she can kid herself that “well, maybe he’d be willing to adopt? maybe he’d be willing to accept a sperm donor”. And he’s telling himself “well, she knows I can’t have kids — the issue is settled — so I guess in a year or so she’ll accept the way things are and move on”.

      I’m not sure what the poker discussion brings to the table, but I don’t play poker and am not interested in it, so I’ll defer to Keith on that.
      However I think it is simply wrong to see that these relationship issues are issues of incomplete information. The point is not that you don’t know what your partners wants, or believes, or is like; it’s that you refuse to BELIEVE it and insist that things (which always means the other person, not you — you’re perfect) will change.

  2. Seems like the wrong poker metaphor. Those couples are holding a pair and wondering whether to stay in the game to draw to a full house.

    1. That presumes that they haven’t already drawn (and that the game is 5-card draw — given the hands, it can’t be Hold-‘Em or Omaha).

  3. In these sorts of relationships, it hurts like hell to “fold” because so much is good about it and because you wish against reason that it can become a clear winner. But in hanging on you are just setting yourself up for an even more horrible loss down the road.

    I know.
    I know.

    But I am still voting for Mr. Obama….

  4. Sometimes hesitation arises from “folding” your hand in a relationship from not knowing when, if ever, you’ll be dealt another worth playing. Bird-hand-bush. Pan-fire. Grass-greener-fence. Devil-you-know.

    1. Of course SCIENCE(TM) can help!
      The basic idea is to accept that you are, essentially, sampling from a population, there is a score associated with each draw from the population, and you want to maximize the value of your score. It is then possible (modulo some assumptions about the population and about how “random” your sampling is) to calculate your optimal strategy, which as I recall is something like: sample some number (I think it’s twelve or so) partners to get a reasonable feel for the range of “scores” in the population, and then sample until you hit a better score than the highest of your twelve.

      You could argue this is a facetious argument, a nice problem for a statistics textbook and nothing more. But I think the very reason it is dismissed is the kernel of wisdom in it. We tend to believe that every person is unique, and specifically that the person simultaneously making us happy and miserable is the only person in the world for us; and this folk psychology is repeated every day in our books, movies, songs. But the fact is that surveys, interviews, looking at the lifetime experiences of most people, etc, all show the same thing — partners are substantially replaceable. Your partner leaves, you are miserable for some time, then you find someone else. And your level of happiness with the new person is, most likely, going to be much the same as it was with the old person.
      WITH THAT IN MIND, one can then realize that
      – if the current level of happiness is not that high, for exogenous reasons, then, yes, chances are you’ll find, in time, someone who gives you the thrills of the current person without the beatings and abuse — or who will please you just as much as the current woman, but without the intense desire for kids
      – if the current level of happiness is basically OK, and you can’t point to anything specific that your partner does that is unacceptable, then you should probably stick with what you have. Giving him/her up means a period of being alone and a period of frustrating search, with little guarantee of an improvement after all that. The problem is probably within yourself and is best dealt with by trying to change yourself as best you can.

    1. Hi Danielle– thanks for the comment. My eyes are lousy also; don’t know how to change font size but will pass on your comment to our tech people.

  5. For the full house, the odds of any of the three other players beating you are at most (3 x (4 royal flushes + 36 straight flushes + 624 fours) / 2,598,960 ) = 0.0007664, or one in 130,480. (I leave out the complications of cards already played and the hands excluded by the full house itself). So it’s the right decision to go all in, even if it goes pear-shaped. The lifetime odds of a relationship going sour are far, far, higher: one in three for marriage in the US, presumably higher for other ties.

    The game of life, the human condition, includes the possibility of tragedy. Not playing isn’t an option. Rationality is not an infallible protection either: that’s the true meaning of the Oedipus story, not the fantasy woven round it by Sigmund Freud – I surmise precisely to conceal a truth as dangerous to his rationalist theory as to its competitors.

  6. I think the major flaw in the analogy is that poker is a zero sum game and life (relationships) is not.

    But at least the poker analysis is correct. The worst hand to hold is the one that is second best when the best hand has the nuts. No way to get that hand to fold.

  7. Bad analogy, because poker isn’t a team game – relationships are. If you’ve turned a relationship into a poker game, there’s no relationship to discuss.

  8. Geeezzz, Keith presents an apt metaphor that illustrates a limited aspect of decision making and how one can make what is ultimately a poor decision, led by a situation with many positive considerations. The poker metaphor maps onto the relationship situation in a limited way, as any metaphor does.
    The criticisms based on those aspects of poker which do not map up the problems of a relationship are bizarre. A metaphor is only a *limited* mapping from one domain to the other (and assumes a similarity in experience). Little else. Consideration of whether or not poker is a team sport, or whether or not the probabilities involved in each situation are similar, are completely irrelevant here, as is the issue of an information theoretic analysis. The only reason I can imagine for bringing in these issues is to demonstrate the commenter’s superior knowledge of probability or how how poker is played.

    It would seem more useful to consider how this and other similar metaphors can be applied to decisions, whether they be about choices in relationships or decisions about politics and politicians. An understanding of the necessary felicity conditions for a metaphor to be successful would be far more interesting. Or what makes a metaphor apt and why is one banal. In any case Keith, an excellent metaphor for dealing with thorny relationship decisions. Thanks.

  9. Isn't it fascinating how we can relate real life situations in a poker game, I've been playing poker since 2010 and had learned a lot ever-since specially with the help of several poker softwares.

Comments are closed.