I was reading Thinking, Fast and Slow by Daniel Kahneman, Nobel Prize winner in Economics. The part of the book that inspired this article was called “Regression to the Mean.” In this chapter, Kahneman recalled something he submitted when he was one of several scientists requested to report their “favorite equation.” Kahneman reported:
Success = Talent + Luck
Great Success = A Little More Talent + A Lot of Luck
Talent is something that stays with you. Luck, not so much. Let’s say a video poker player had the following annual results:
| 2009: +5,000 |
| 2010: +9,000 |
| 2011: +7,000 |
| 2012: +33,000 |
What would your prediction about what their score will be in 2013, assuming you were making this prediction on January 1, 2013?
It’s clear that this player has some talent. There is an upward trend where over time the scores for this player are getting better, although 2011 was a little worse than 2010. Most players lose at video poker and winning the past four years in a row is a better-than-average result. The 2012 score is quite a bit larger than the others. Do you think that kind of success will probably continue or do you think the score will revert to the lower level of success?
If I had to pick a number I’d probably pick something in the range of $16,000 as my best guess. That’s the average of the four numbers. I think the breakout year was probably an anomaly and the score in 2013 will be lower than the maximum. I believe most people would pick a number quite a bit higher than I would. They would “spot a trend” and go with it.
You see this in sports all the time. Players have a breakout year — get a huge contract — and then become not-so-spectacular. Sports writers make up all sorts of explanations on ‘why’ this happens, but the simple explanation that makes sense is that during the breakout year, this player got extra lucky. If he’s a baseball player, the balls he hit tended to be barely fair rather than barely foul. His hit balls “had eyes” and found their way between the defenders. On the day he faced the league’s best pitcher, that pitcher was having an off day. He didn’t get injured and didn’t face stressful situations off the field. Good luck is not something you can count on over time. It’s very unlikely that this kind of luck will happen next year as well.
Back when my “Million Dollar Video Poker” autobiography first came out where I wrote that I won $1,000,000 over six months, one of my professional player friends commented that all the book showed was that I got lucky over that half-year. I was offended by his assessment, but now I think there was a lot of truth to what he was saying. Yes I had some talent to recognize and capitalize on some situations, but being in the right place at the right time and being greatly over-royaled for six months accounted for much of my success during that million-dollar run. Although I’ve almost always had positive annual scores (which indicates a certain level of talent) I’ve never again come close to earning a million dollars in one year (which indicates that much of that result was due to luck rather than skill.)
The same thing works in reverse. If you had your biggest loss ever last year, this year will probably be better. The same as good luck can’t be counted on, neither can bad luck. Your worst-ever result in any year is probably more due to bad luck than lack of skill. You pick situations carefully and make the correct plays, but sometimes the good hands come and sometimes they don’t.
You hear players making explanations/excuses as why they had their biggest loss recently — and this explanation usually involves the casinos or game manufacturers tightening the machines or perhaps the promotions not being so generous. The truth is that machines are generally tighter than they were a year ago and the promotions are less generous but this doesn’t matter very much. Good situations can still be found. Good players can find those good situations. The best players are still winning and the not-so-good players are still making excuses.
The real explanation, however, for those who had their worst year last year is that they were relatively unlucky compared to what happened in other years.
In my Secrets of a Video Poker class (next occurring Tuesday April 9 at the South Point), I talk about luck versus skill in video poker. I say that over the next two hours luck accounts for probably 80% of your results. Over the next 200 hours, luck accounts for probably 20%. While this is generally true, Kahneman’s book is causing me to rethink the subject. I now believe that it’s a little more complicated than what I mention in that class.
Let’s look again at professional baseball. I’m from Los Angeles originally and I like both the Angels and Dodgers to do well. Both should be better-than-average this year and them meeting in the World Series is definitely possible (albeit not very likely.) In that best-of-seven series, luck will play a major role in who wins.
If I get a bunch of video poker players from the South Point to form a team and we go up against either the Angels or Dodgers, luck will not have anything to do with it. We’ll get slaughtered — 100-0 — and that’s just in the first inning.
The same is true in video poker. Among players, highly skilled or not, there will always be a luck component in the results. But among highly skilled players, the luck will cause the annual results to vary primarily in positive territory. With not-so-highly skilled players, luck will cause the annual results to vary in negative territory.
Improving your skill DOES help your results on average. Improving your skill will NEVER eliminate the variance in your score. Exactly where your score ends up will always have a luck component in it — but over several years where your score ends up will have a very large skill component in it as well.