“This one paid for the second one, that won me $20, so I bought the third one. That one bought one more, and I won $40.” I’d asked my friend for a pen, but as she dug through her purse she felt the need to explain the lottery tickets in there. “As soon as I cash these in,” she continued, “I’m not buying any more. I’m not about to press my luck.”
Not pressing your luck when you’re gambling is probably a good thing. They call the lottery “The Poor Man’s Tax” for a reason. The lavish décor and free drinks at casinos should be a clue. The house wins quite a bit more than its guests. But even when we’re not gambling, we have an inner thermostat that tells us when to stop. From childhood to our twenties and maybe later, we work hard to identify how many times we should try for something. When we “win,” we feel smart. When the future doesn’t work out the way we anticipated, we adjust our thermostat. This thermostat is a heuristic, or mental shortcut, programmed into us for survival. Primitive man thought, “A little closer to the bear’s cave, and I’ll have enough berries. After that, I’d better go before she smells me.” Young me said, “Dad, can I have this Barbie®?” “Dad, look how pretty she is.” “Dad, she only costs $9.99.” Then I knew, “If I ask one more time, Dad will get mad.” I wanted it sooooo badly, but I learned to exercise restraint. My thermostat kept things cool.
Heuristics are primitive tools. They can be helpful, and they feel right, but they aren’t based on logic. You and I want to be wiser than Captain Caveman, so we should learn to recognize when the thermostat is just blowing hot air. When you suddenly change direction, are you responding to a verifiable danger? The bear can smell you outside her cave. Are you responding to the belief that if something positive has happened recently, it’s not likely to continue? That’s the Gambler’s Fallacy. Many casinos list the most recent 20 spins of the roulette wheel. If the last 4 spins have landed on black, betters often switch to red. If out of 20 spins, most have been black, and the last spin was red, betters might switch to black. In actuality, the wheel is not programmed to hit any quantity of reds or blacks before switching direction. There is no statistic or relevant influence which backs up these bets.
So you don’t gamble? Smart. Have you ever met a couple who says, “We’ve had 2 boys, so we’re due for a girl.” Millions of boys have been born because of this faulty logic.
The Gambler’s Fallacy is subtle. As I’ve researched and written the past 500 words, I can’t tell you how many times my heater has kicked on. It’s a chilly February day, but as long as I feel warm, I trust my thermostat to keep me that way. We do the same thing with decision making, and it can have some serious consequences.
In a 2015 study, Kole, Kanz and Klapper gave loan officers some applications which had previously been approved or declined. The reviewing officers didn’t know the previous decisions, but were asked to approve or decline them according to the same rules. There were two differences between the initial and second review of these applications: 1) The loan applications were shuffled, so they were handled in a different order the second time. 2) The officers were paid on three incentive schemes, which resulted in different levels of accuracy. It turns out up to 9% of the loan decisions had been made based on the order in which the cases appeared. The facts were the same, but a loan officer would decline an application after approving a certain number of loans, even if the declined application met the same criteria as the others. The Gambler’s Fallacy didn’t have as much impact when officers were paid based on accuracy and given more time to review each loan.
In 2016, Daniel Chen, Tobias Moskowitz and Kelly Shue reviewed immigration cases, where refugees requested asylum due to persecution or danger in their home countries. They found that the order of cases had an influence on the decisions of the judges. If a judge approved a case, the next case was up to 3.3% more likely to be denied. Allowing for other variables, these researchers estimate that about 2% of the denials were based purely on the order in which they were seen. Out of 150,357 decisions reviewed, that’s 3,007 people whose lives are endangered because of a primitive response mechanism.
Our decisions matter. Maybe you’re hiring job applicants, deciding whether to help the guy with the sign on the exit ramp, choosing whether to speak up even if you risk upsetting someone. Since the thermostat is so subtle, it’s easy to miss. The research shows these steps can help:
- Use measurable rules for your decisions. Jeremiah 17:9 ESV says, “The heart is deceitful above all things.” Call it your heart, your gut or heuristics, this mystery organ is shockingly unreliable. Re-evaluate your rules periodically. Have your choices proven to be right?
- Take time to examine each case as though it’s the first.
- Step away between cases. Taking a break can help you disconnect the new case from the old.
- Shuffle the deck and ask someone else review your work. Humility improves accuracy.
- Check your motivation. Sometimes we switch direction out of fear that others will see us as too lenient or too harsh. Instead, make choices based on verifiable data.
- Examine other things which might be influencing your choices. The asylum cases were drawn from a variety of judges. Some judges had a record of approving only 10% of their cases, while others approved 80% (they considered this in their research technique). The system doesn’t assign only bad cases to one judge and good cases to another. Clearly, the judges had other biases when making their decisions.
Next week, we’ll continue looking at faulty thinking and how we can use wisdom rather than automation. If you missed last week’s discussion on Anchoring, check it out here.