Tuesday, December 4, 2012

Playing On a Bell Curve

Today I was drawing a bit of a blank on what to talk about, which is interesting, because even saying that made me think of a new topic.  The topic is based on a concept called "regression to the mean," which is another way of saying, "things tend to return to normal after awhile."

Your performance in anything can typically be represented by a bell-curve.  You have a general level of performance, your most common middle ground, and you will play to that level most of the time.  Sometimes you will have your off days, sometimes you will have your on days.  Probability states that, your performance will fall somewhere in the middle.  That means you typically can't rely on playing your "best."

There's another bias people are susceptible to, which is the "optimism bias."  When you ask people to plan things out, they tend to assume everything will go right.  In fact, when you ask them to plan things out normally, then ask them to plan things as if everything will work perfectly, the plans tend to look the same.

Let's combine these concepts.  You probably have an idea in your head about your "real" skill level.  And this skill level, I'm willing to bet, looks a lot like you playing at your absolute best because of the optimism bias.  The thing is, that "best play" is something of an anomaly.  Your "real" skill level, what you can bank on happening a majority of the time, is going to be worse than that.

So the result?  You're normally going to be disappointed with how you play!  You may optimistically assume everything will fall into place, and anything less than that is "bad."  So it's no surprise that you're almost always "off" when it counts, because you're defining your skill in a way that guarantees it to be the case.  Considering how stress and disappointment negatively affect the way you play, it's important to be honestly pessimistic.  You are probably going to play at your average, and then if you don't, you've got a coinflip's chance of being above or below it.

But how can you use this kind of knowledge to your advantage?  First, it'll help you feel less bad when you aren't playing your absolute best, and that should help keep you from slipping down to your worst.  Second, it should give you hope that when you are performing badly, odds are you can ride it out and things will get better on their own.  Third, it tells you what your real goal is; it's not to always bring your A-game (generally impossible) but to bring your B-game up as high as you can get it.  Practice your basics, the stuff that's as un-screwup-able as can be.

Something else to keep in mind: I don't think that your performance is some mystical thing you don't control (there wouldn't be much point to this blog, if it were).  It's not entirely up to chance whether you have a bad day or not; there are steps you can take to make it more likely you will play on the upper end of whatever skill curve you possess.  I'll be going into that stuff on Friday.  Though they differ from person to person, I'll try and stick to ones that have definitely worked for me, or ones that just have a basic grounding biology.

Thanks for reading.


  1. Richard "Delphiki" CappelDecember 27, 2013 at 2:14 AM

    This is a great thing to point out. Sometimes I'll think "I could beat so-and-so if I was on my game", that is if I performed as well as I knew HOW to. It's one thing to know better, and another to do better.
    This reminds me of a bit in Daniel Kahneman's book in which he describes his experience assisting in pilot training programs for the Israeli air force. What an instructor thought was the effect of his efforts at discipline was actually the result of this kind of bell curve effect - when you yell at the pilot who went too fast, he's going to go slower next time, but he would have anyway, because he performed so exceptionally the first time. The outlier will give way to averages. The same thing occurs when trying to chastise an exceptionally poor performance, and that opens up the possibility for a ton of attribution mistakes.
    Finally, I would suggest that being disappointed in your average performances actually has a survival benefit, because it indirectly highlights those higher-than-average performances by chalking up such a high reward to whatever "good trick" we might have attributed the reward to. The idea that we're usually going to be disappointed in average performances is something I'm definitely going to try and keep in mind.
    This article is very concise; I like that. The second- and third-to-last paragraphs just shine!

  2. There's a neat Hunter X Hunter dialog scene / chart that illustrates this concept as well. Let's see if I can find it.... I can't.