Superbowl betting on the coin flip

By Marc Resnick

There are two competing decision making biases/heuristics that are pretty common.  If you flip a coin five times and it comes up heads every time (which even a fair coin will do once in a while), there are some people that will estimate the chance of another head as higher than 50% because it is on a “streak.”  There are also people with the opposite bias who would think that a tail is “due” so is more likely.  But among bettors, belief in streaks is a stronger tendency than a belief in reversion to the mean.  So for the Superbowl, if there has been a streak of heads or tails, there are more people who bet on the streak continuing than the opposite.  The gambling houses take advantage of this by adjusting the odds so that they get more losers than winners. Pretty smart use of human decision making insights to separate a fool from his/her money.

How can we use this insight on the shop floor?  What if we take advantage of people seeing streaks by motivating superior performance based on past success?  Even when there is no factual basis for it, we can get superior performance from employees just because they think that they are on a streak.  But when past performance is bad, we can focus on them being “due” for a good performance.  As with many things in life, confirmation bias is often a self-fulfilling prophecy.  The better you think you are going to do, the better you actually do.  We can use both sides of the coin to prime the performance we are looking for.

1 Comment so far

  1. Sharon


    I like this idea. I’d like to add that the difference between a coin flip and employee performance is that the coin flips are independent. That’s why “streaks” and “being due” are erroneous. But I’d argue that events involving success on the job are not independent. I think they are positively correlated, and I use this to motivate my colleagues. Though I guess one could argue that they are negatively correlated. In any case, I think this makes your argument stronger.


Comments are closed.