Anyone that follows basketball knows about the “hot hand.” That’s when someone gets on a roll, making several baskets in a row. Teammates pick up on it and start feeding him the ball. Defenders start double teaming. The audience gets louder with each consecutive basket, and everyone expects him to make his next shot.
Yet analysis of thousands of shot sequences prove that there is no such thing as a “hot hand.”
Sure, some players are better shots than others, and make a higher percentage of their shots. But “hot hands” aren’t, actually. It’s our minds trying to see patterns that aren’t really there. NBA players often try to ride a 3-point streak, but most actually have a worse percentage shooting after making one than missing one, according to Scientific American.
The Gates Foundation wanted to know what the most successful schools in the country had in common.
Researchers found that in the top 50 schools in Pennsylvania, there were 4 times more small schools than were expected. Everyone agreed that smaller schools would mean smaller classes and more personal attention for the students, so the Gates Foundation invested $1.7 billion in small schools, sometimes even splitting large schools into smaller ones.
But those school researchers never asked about the characteristics of the worst schools, which were also smaller than average. Turns out that smaller schools weren’t better or worse, they were just more variable. When you look at a bigger sampling, larger schools actually provide better results, and more curricular options as grades get higher.
Both of these examples are simply mistakes of using small samples.
According to Daniel Kahneman, Nobel Laureate, in his book Thinking Fast and Slow: “Statistics produce many observations that appear to beg for causal explanations but do not lend themselves to such explanations. Many facts of the world are due to chance, including accidents of sampling. And causal explanations of chance events are inevitably wrong.”
For those of us that are not Nobel Laureates, let me try to put it another way:
Be wary of easy explanations and small numbers.
Life has a lot of chaos, so our minds naturally want to make sense out of all the randomness. As a result, we are too quick to see patterns out of a group of facts or sequences that are simply too small to be accurate.
Focus groups? Random member interviews? Think of them as good sources of information, but not enough on which to base major decisions. You need a large number of results to be able to verify accurate results, otherwise you are simply following a trend, a “hot hand.”
- To become a CEO, think like a CEO - August 11, 2021
- Data Cooperative: Big Crazy or Not-So-Crazy Idea for CUs? - June 18, 2021
- Why should I care about social responsibility? - April 20, 2021