You May be Better than You Think: How to Evaluate Your Performance

When evaluating your performance at sports, at school or at work, you often pay too little attention to the original probability of reaching this level of performance. Your performance may be better than you think!

Golden Behaviors

– Nudges for a Healthy Lifestyle –

Reading time: 6 minutes

Playing squash again

Well, I’m back at playing squash: what a relief! It was fun to run, although it is not the most fun activity I can think of, I let behavioral science work in my favor. So, I am glad to be back at playing squash. I play 3-4 times a week, working on improving my game. However, no matter how much I practice, I do not seem to improve.

When playing my team mates I still lose from the players that I usually lose from. I still win from the players I usually win from. It seems that no matter how much I play, I cannot consistently improve my game compared to the better players in my team. It also seems, that no matter how often my team members play, they do not seem to be getting better either. So, is this endless practicing utterly useless?

Reality is that practicing really helps to improve your game or other activity. However, we usually fall prey to the base rate fallacy, that makes us believe that we are not improving, whereas in reality we are. Comparing myself to my team mates may not be the most accurate comparison to evaluate whether I am improving.

Base rate fallacy explained

The base rate fallacy is the tendency of people to give the original rate of possibility too little attention. The Nobel Prize winner Daniel Kahneman provides a good example of the base rate fallacy in his book Thinking Fast and Slow (2011). Consider the following story and say out loud your intuitive answer to the question:

• An accident occurs somewhere in a city.
• There are two color taxis in the city: 85% green taxis and 15% blue taxis.
• A witness has seen that it was a blue taxi. Reliability of the witness is 80% correct and 20% incorrect.

What is the probability that the taxi was really blue?

If you are like most people you base your answer on the fact that the reliability of the witness is 80% and hence, the probability of the taxi being blue is 80%. Is this what you have done as well?

In that case, you have ignored the proportion of falsely identifying a green taxi as blue. After all, in 20% of the cases the witness will (falsely) identify a green taxi as blue. In case of 100 accidents this would be 17 green taxis (falsely) identified as blue (20% x 85 green taxi’s). You should compare this to the proportion of correctly identifying a green taxi as green. In case of 100 accidents this would be 12 blue taxis (correctly) identified as blue (80% x 15 blue taxi’s). Thus, for the 29 taxis that the witness would identify as being blue, only 12 will be really blue. Thus the probability that the taxi was really blue is about 41% (12 divided by 29).

Drunk driving

Now that you know what to look for, it may seem easy. In that case, try this one:

• Let’s assume, one in a thousand drivers is driving drunk.
• Breathalyzers display false drunkenness in 5% of the cases.
• A police officer stops a driver at random and administers a breath test.
• The breathalyzer gives a positive test result for drunk driving

What is the probability that this driver is really drunk?

So what do you think? Say out loud your intuitive answer to the question. Look for the answer at the bottom of this blog.

What is my base rate?

Comparing myself to my team mates is most easily, because I play them every week. This makes it easy to compare, but then I overlook the fact that we are both practicing and are both improving our game. I do not take into account the group of people who do not improve. It may be difficult to estimate the size of this group: it could be tested in the scenario of the taxi’s by testing the reliability of the witness, but it has to be assumed in scenario of the breathalyzer. This occurs with many social comparisons of behavior, such as comparison in performance at sports, school or at work.

There are two main reasons for the base rate fallacy to occur. First, it takes hardly any effort to compare yourself to your own social group – as I did comparing myself with my squash team members. Second, it may be difficult to find a more objective measure to which you can evaluate your performance. This is because you meet many people who are working on improving their performance, but few people who are not, because they do not show up at your sports club, school or training.

Practice to move up the ladder

Instead of comparing myself to my team mates, I could compare myself to a larger group of squash players, for example, all the players from my squash club. I could participate in a concept you may be familiar with: the squash ladder. You start at the bottom of the ladder and when you win from  players that rank higher than you, you take their place on the ladder.

Although this is a larger group for comparison, this comparison is not without bias. After all, not all squash players in my squash club participate on the ladder and players that do participate probably play more often than the average squash player. Because these other players practice a lot – just like me – they improve as well. This is what I observe at the squash ladder of about 30 people at my squash club: my position is relatively stable throughout the season.

Comparison to larger groups are more objective

If you are a player participating in the competition, than you have the opportunity to compare yourself with an even larger group of players. You can compare your ranking to all Dutch competition players. In squash competition there is a rather advanced points awarding system to monitor your performance. When I look at the last competition season, I grew from about 1350 to almost 1550 points (11% change), which seems like good improvement.

Although this is an even larger group for comparison, this comparison is still not without bias. After all, not all squash players participate in the competition. Again, competition players probably play more often than the average squash player. Therefore, this is still not an objective evaluation of my improvement, but rather a relative evaluation compared to a selective group of others.

Drunk driving revisited

So did you think of an answer to the question above: “What is the probability that this driver is really drunk?” If you haven’t already done so, say out loud your intuitive answer to this question. Did you take into account the false positive test results? In this specific scenario where we assume that police officers stop drivers truly at random, and not because of bad driving. So the probability that a driver is really drunk is about two percent. So let’s hope that the reliability of the breathalyzer is better than the 5% of false drunkenness in the test.

Evaluating your performance at sports, school or work

So, how should you evaluate our performance at sports, school or work?

1. First, comparison to larger groups are usually more objective. So try to compare yourself to a group that is larger than the group of people playing at your local sports club, the school you are attending, or the employees in your specific company.
2. Second, most social comparisons are relative and say little about your objective performance. Although relative comparison is increasingly important in all areas of our live, look at the positive side when you have not improved you relative ranking. It says little about your objective improvement.
3. Third, and maybe most important: if you practice, you will get better. As we have seen, it can be very difficult to evaluate performance in an objective and absolute way. So the most important thing to remember is: if you practice, you will get better. You can be sure of it.

Niels Vink (1975) is author of Golden Behaviors and behavioral designer. He uses insights from the behavioral sciences to explain why people often act against their own interests. As a behavioral expert, he explores how you can nudge your behavior for a healthy lifestyle. He has Master degrees in Social Psychology (Leiden University) and Industrial Design Engineering (Delft University of Technology) and holds a PhD in Consumer Behavior.

When you have been inspired to start and maintain your Golden Behaviors, reach out to me.

SUBSCRIBE TO OUR BLOG

Golden Behaviors Blog

Name*

Email*