#79 Using Intuition To Make Important Decisions: Don't Foget The Statistics

Many years ago an employee made frivolous claims against some folks in one of our offices. I thought the whole thing was BS and was ready to fight it out.

Legal counsel disagreed. 

"Write a check", they said. “The other side will sue, they'll slow you down, legal fees will add up fast. Bite your tongue, write the check, and move on."


We’ll get back to this threatened lawsuit in a bit. But first I want to give the discussion some context. I want to expand on how we use our intuition to make these kinds of decisions.

Our brains, sometimes consciously but more often unconsciously, assign weights to different outcomes.

There’s a 40% chance she’ll accept the job offer.
There’s an 80% chance I can get through that yellow light before it turns red.
There’s a 95% chance we’ll win the lawsuit.

These are called decision weights, and, unfortunately, we tend to intuitively suck at assigning them. The only time we get them right are in extreme cases.

These conclusions are per Nobel Prize winning psychologist Daniel Kahneman. In his book, Thinking Fast and Slow, he provides the following chart on page 315.


It shows the real probability of a gain (top row) versus the decision weight we assign (bottom row).

The blue ovals show the rare occasions where we intuitively assign decision weights in sync with probability. Everywhere else on the chart our intuitive decision weights are faulty. And our errors are especially interesting in the areas I've circled in green and red.

Let’s start with green. Notice that when we have a 2% chance of a gain our brain tells us we have an 8.1% chance of a gain. We're overweighting our odds of good fortune by 4 times the proper value.

Point being, we're suckers for a long shot. Which explains the existence of the lottery. Even though we’re more likely to be struck by lightning WHILE drowning than to hold the winning ticket, we still plunk down our money.

This tendency is what Kahneman calls the “Possibility Effect”. Meaning highly unlikely outcomes get much higher weightings than they should.

On the other end of the diagram is the red circle. When we have a 99% chance of a gain it only feels like we have a 91.2% chance of a gain. This is why folks settle lawsuits for less than they would likely be awarded.

Per the chart, if you had a 99% chance of winning a $20 million lawsuit you would probably accept an $18.24 ( 20 X .912) million dollar settlement prior to the judges ruling. That 1% chance of NOT getting the big payday feels like an 8.8% chance (100 - 91.2 = 8.8). So you leave $1.76 million (20 - 18.24 = 1.76) on the table. 

This tendency is what Kahneman calls the “Certainty Effect”. Meaning highly likely outcomes get much lower decision weights than they should.


Now that I’ve given you the official names - Certainty and Possibility Effects - let’s ignore them. Lets instead simplify things by thinking in terms of being RISK SEEKING or RISK AVERSE. And let's look at our appetite for risk across these 4 scenarios.

Scenario A - High Probability of a Significant Gain
Scenario B - High Probability of a Significant Loss
Scenario C - Low Probability of a Significant Gain
Scenario D - Low Probability of a Significant Loss

In my brain I tend to equate the high probability of a Gain with a low probability of a Loss. They’re not the same thing, but emotionally they feel similar to me. And, assuming we’re talking about dollar values large enough to get our attention, my gut feel is correct.  These two scenarios…

A - High Probability of a Gain - 95% chance to win $10,000
D - Low Probability of a Loss - 5% chance to lose $10,000

…both lead to an intuitive feeling of Risk Aversion or playing it safe. So in scenario A, a high probability of a gain, I would accept a lesser payout if it were guaranteed. So I might accept $9,000 up front rather than find out if I really got the $10,000. It’s a bad decision statistically, but there's a big difference between statistics and peace of mind. What's it worth to sleep well at night?

Scenario D - low probability of a significant loss - is another example of RISK AVERSION. And we all write checks for this one every year. The odds of my house burning down are close to zero. But I pay big bucks for insurance that covers me just in case.

If it weren’t for this supposedly irrational thinking the insurance industry wouldn’t exist. But again, peace of mind and statistics are two different things.

Now let’s looks at the other two scenarios.


As a reminder, here are the other two scenarios.

B - High Probability of a Loss - 95% chance to lose $10,000
C - Low Probability of a Gain - 5% chance to win $10,000

These two tend to make us RISK SEEKERS, willing to roll the dice so to speak. Scenario C, Low Probability of a Significant Gain, is the lottery example we already talked about. Scenario B is one we don’t think about very often, and the most surprising finding from Kahneman’s work. 

Say you’re being sued and it’s not looking good. There’s a 95% chance you’re going to lose ten grand. If you’re into statistics it would be said that you have an expected loss of $9,500 ($10,000 x 95%). So if the other party offered to drop the suit if you wrote them a check for $9,000 you should say yes. Negative $9,000 is better than your expected outcome of negative $9,500. But most folks would say no. We HATE locking in losses so we’re more likely to roll the dice and pray for that 5% chance of winning. We become RISK SEEKERS when the odds are stacked against us.

This seems odd, you would expect us to want to protect our downside when we're up against the wall. But, instead, we intuitively decide to swing for the fences.


Now that we’ve looked at these scenarios, let’s take a look at how the lawsuit I mentioned in the opening played out. Let’s see if we ended up playing the game as Kahneman would expect.

I, having zippo courtroom experience, felt there was a zero percent chance we could lose the case. We hadn't done anything wrong and the justice system is there to protect wonderful folks like us. In retrospect, quite naive. But regardless, I thought we were dealing with certainty.


Our experienced lawyers knew that we were actually in scenario D - low probability of a big loss. Yes, we would almost certainly win, but there's nothing certain in a courtroom. So they were intuitively feeling extremely RISK AVERSE. And willing to pay well beyond expected value to get rid of the risk. Just as Kahneman would expect.

The sleazebag lawyers representing the employee were in scenario C - low probability of a big gain. They were playing the lottery - RISK SEEKING. Again, just as Kahneman would expect.

So what happened? 

It's pretty easy to guess. Us RISK AVERSE, they RISK SEEKING. We paid out way more than we should, tens of thousands of dollars more, to avoid the risk.

It sucks, but that's how the story ends. And I guess that's how these sleazebag attorneys afford all those billboards and TV commercials. The game, as Kahneman proves, is intuitively slanted in their emotional favor.


Whether you’re buried in a lawsuit or are considering a critical partnership, there’s no downside to this kind of analysis. 

Where do you stand on the risk spectrum?
Where does the other party stand?
Are the Certainty and Possibility Effects leading you down a suboptimal path?
Would you be better off being more statistically logical and less intuitive?


And Kahneman has a strong opinion on that last question.

“Consistent overweighting of improbable outcomes - a feature of intuitive decision making - eventually leads to inferior outcomes.”

But, here’s a case where I think you should be careful. Yes, statistically you should always go with the expected value. That’s the right answer in the long run when you get many trials. But we don’t always get lots of trials and we don’t have unlimited cash or time in the short run. Maybe we only get one or two chances at major strategic decisions - sign the partnership agreement or don’t, accept the buyout offer or don't, settle the lawsuit or don't. 

So with a tiny number of trials, maybe a blended process makes sense. Maybe statistics and intuition can work together to give us a better answer than either can working alone.


(This site is all about building a Map that will help me do work and life better. So at the end of each post I check in to see if any changes / insights come to mind.)

It’s WRONG STORIES again. We’re not good at intuitively assigning decision weights and we need to keep this fact front and center in our minds when faced with big decisions.


Daniel Kahneman’s book - Thinking Fast and Slow - is one of my favorite books. He does a great job of explaining how we come to wrong or biased conclusions - WRONG STORIES. And that’s why I’m doing this series of posts covering the topics that Kahneman writes about. Here’s a list of the prior posts in the series.

# 73 - Being A Jerk Seems To Work
# 74 - A Well Timed Pizza Could Have Changed Old Red’s Life
# 75 - Does Thinking About Money Mess You Up
# 76 - Luck Can Take You To The Top But It Won't Keep You There
# 77 - Do Not Take Business Advice From Dart Throwing Monkeys
# 78 - Is Following This Site Worth Your Time


I hope you enjoy them.

***Note: This site works best when you read the posts in order. So please head to the ARCHIVE to get started.