Latest / The Dutch Investors / #41 | Mental models every investor should know | Causation vs Correlation | Part 3

Transcript

#41 | Mental models every investor should know | Causation vs Correlation | Part 3 · The Dutch Investors

• 0:00
  SPEAKER_00: Have you ever noticed that when the church bell rings, the dog across the street starts barking?
• 0:06
  SPEAKER_00: Happens once and you don't think much of it.
• 0:09
  SPEAKER_00: Happens twice, three times, you start to wonder, is the bell making the dog bark?
• 0:15
  SPEAKER_00: But what if the real answer is something else entirely?
• 0:19
  SPEAKER_00: Today we're talking about one of the most misleading tricks our brain plays on us: confusing correlation with causation.
• 0:27
  SPEAKER_00: It sounds like a technicality, but it shows up in almost every corner of our life, from headlines to marketing to investing and everyday decisions.
• 0:38
  SPEAKER_00: So let's talk mental models part 3.
• 0:41
  SPEAKER_00: Causation vs.
• 0:42
  SPEAKER_00: Correlation.
• 0:52
  SPEAKER_03: Could anyone do the same thing you did?
• 0:57
  SPEAKER_02: Are you born with it?
• 0:58
  SPEAKER_02: Is it DNA?
• 1:00
  SPEAKER_02: Is it some early childhood exposure?
• 1:04
  SPEAKER_03: The first moment of investment is don't lose.
• 1:06
  SPEAKER_03: And the second form of investment is don't forget the first rule, and that's all the rules are up.
• 1:15
  SPEAKER_03: Just doesn't make any sense to us.
• 1:16
  SPEAKER_03: Mommy, we are going to be buyers of things over time.
• 1:19
  SPEAKER_01: What I needed to get ahead was to greet meet against idiots.
• 1:23
  SPEAKER_01: And luckily there's a large supply.
• 1:27
  SPEAKER_00: Let's say you're walking through a small town, and every time the church bell rings, a dog starts barking across the street.
• 1:35
  SPEAKER_00: It happens once, then again.
• 1:38
  SPEAKER_00: You pass through the next day, same thing.
• 1:42
  SPEAKER_00: The bell rings, the dog barks.
• 1:45
  SPEAKER_00: Now a part of you, the pattern-seeking, story-loving part of your brain, starts to wonder and ask questions.
• 1:53
  SPEAKER_00: Is the bell making the dog bark?
• 1:56
  SPEAKER_00: It seems like it, right?
• 1:58
  SPEAKER_00: But step back for a moment.
• 2:01
  SPEAKER_00: Maybe the bell rings at noon.
• 2:03
  SPEAKER_00: Maybe the mailman comes at noon.
• 2:05
  SPEAKER_00: And maybe the dog just really hates the mailman.
• 2:09
  SPEAKER_00: This is what we're talking about when we explore the difference between correlation and causation.
• 2:14
  SPEAKER_00: Because it's one of the most important mental models you can carry around in your pocket, because mistaking one for the other, assuming one thing causes another, just because they happen together, leads to some of the worst decisions we make in life, business, investing, and even in science.
• 2:33
  SPEAKER_00: I love the following example by Daniel Kahneman from the excellent book Thinking Fast and Slow.
• 2:39
  SPEAKER_00: He wrote, and I quote, Depressed children treated with an energy drink improve significantly over a three-month period.
• 2:48
  SPEAKER_00: I made up this newspaper headline, but the fact it reports is true.
• 2:52
  SPEAKER_00: If you treated a group of depressed children for some time with an energy drink, they would show a clinically significant improvement.
• 3:01
  SPEAKER_00: It is also the case that depressed children who spend some time standing on their head or hug a cat for 20 minutes a day will also show improvement.
• 3:10
  SPEAKER_00: Whenever coming across such headlines, it is very tempting to jump to conclusions that energy drinks, standing on your head, or hugging cats are perfectly viable cures for depression.
• 3:22
  SPEAKER_00: These cases, however, once again embody regression to the mean.
• 3:27
  SPEAKER_00: I'm quoting Kahneman again here.
• 3:29
  SPEAKER_00: Depressed children are an extreme group.
• 3:32
  SPEAKER_00: They are more depressed than most other children, and extreme groups regress to the mean over time.
• 3:38
  SPEAKER_00: The correlation between depression scores on successive occasions of testing is less than perfect, so there will be regression to the mean.
• 3:47
  SPEAKER_00: Depressed children will get somewhat better over time, even if they hug no cats and drink no Red Bull.
• 3:55
  SPEAKER_00: Alright, so let's break this down for a second.
• 3:57
  SPEAKER_00: Correlation is when two things tend to happen around the same time or in a similar pattern.
• 4:04
  SPEAKER_00: Causation is when one thing directly causes the other to happen.
• 4:10
  SPEAKER_00: And that might sound like a small distinction, but it isn't.
• 4:14
  SPEAKER_00: It's the difference between understanding how the world works and getting tricked by your own brain.
• 4:21
  SPEAKER_00: Because our brains are wired for stories.
• 4:24
  SPEAKER_00: We don't like randomness.
• 4:27
  SPEAKER_00: We want meaning.
• 4:28
  SPEAKER_00: We want to believe that if A happens, then B happens.
• 4:33
  SPEAKER_00: A must have caused B.
• 4:35
  SPEAKER_00: But unfortunately, the universe doesn't always work that way.
• 4:39
  SPEAKER_00: The universe, the stock market, your brain is under no obligation to make sense to you.
• 4:45
  SPEAKER_00: Let me give you an example.
• 4:47
  SPEAKER_00: In the summer months, the ice cream man sells a lot of ice cream, right?
• 4:53
  SPEAKER_00: Makes sense.
• 4:54
  SPEAKER_00: And during winter, they sell a lot less.
• 4:57
  SPEAKER_00: During summer, shark attacks also increase.
• 5:01
  SPEAKER_00: So both ice cream sales and shark attacks increase during summertime.
• 5:07
  SPEAKER_00: Now no reasonable person would argue that ice cream is attracting sharks, right?
• 5:13
  SPEAKER_00: But they are connected.
• 5:14
  SPEAKER_00: I mean, that's what we're saying here.
• 5:17
  SPEAKER_00: They are, but not through each other.
• 5:21
  SPEAKER_00: They're both connected to a third variable.
• 5:24
  SPEAKER_00: Heat.
• 5:25
  SPEAKER_00: It's summer.
• 5:27
  SPEAKER_00: More people go swimming.
• 5:28
  SPEAKER_00: More people want ice cream.
• 5:30
  SPEAKER_00: This is a perfect example of correlation and not causation.
• 5:35
  SPEAKER_00: The two things move together, but one isn't pushing or pulling the other.
• 5:40
  SPEAKER_00: Staticians use something called a correlation coefficient.
• 5:44
  SPEAKER_00: It's a bit technical, but they use a number between minus one and plus one.
• 5:50
  SPEAKER_00: The closer you get to one, the more perfectly two things move together.
• 5:56
  SPEAKER_00: A correlation of one is like Celsius and Fahrenheit.
• 6:00
  SPEAKER_00: One degree of Celsius will always have the same match in Fahrenheit.
• 6:05
  SPEAKER_00: That's a perfect correlation.
• 6:07
  SPEAKER_00: Change one and the other must also change.
• 6:11
  SPEAKER_00: But a correlation of zero is like plotting the average bottled water consumption in 100 countries against their suicide rates.
• 6:18
  SPEAKER_00: There's no real pattern.
• 6:21
  SPEAKER_00: It's scattered.
• 6:22
  SPEAKER_00: Nothing meaningful connects the two variables.
• 6:25
  SPEAKER_00: And yet people still try to find stories in that scatter.
• 6:29
  SPEAKER_00: Still try to claim, because this can actually be the case, that in countries with high bottled water consumption, suicide is more common.
• 6:39
  SPEAKER_00: And although, first of all, most people will take that at face value and drink less bottled water, but the truth is, even though our brains don't like it, it's purely coincidence.
• 6:51
  SPEAKER_00: Imagine a company changes its logo, and two months later, revenue starts growing again.
• 6:58
  SPEAKER_00: Someone in the meeting says, See, the rebranding worked! And just like that, the logo gets credit.
• 7:06
  SPEAKER_00: But no one really dares to ask, hmm, is it really the logo?
• 7:11
  SPEAKER_00: What else changed in those two months?
• 7:14
  SPEAKER_00: Maybe the weather improved, more food traffic came through the doors, maybe a competitor raised prices and lost some customers, maybe the economy picked up just enough to tilt the numbers upwards.
• 7:27
  SPEAKER_00: It's not what our brain likes.
• 7:28
  SPEAKER_00: We want to know that we are in control and we make sure that things happen or don't happen.
• 7:35
  SPEAKER_00: If you assume causation when there's only correlation, you'll make decisions based on ghosts, on imagined forces that aren't there.
• 7:46
  SPEAKER_00: And once you believe a false cause, you'll keep chasing it.
• 7:50
  SPEAKER_00: You'll double down, you'll defend it, and every time the pattern appears again, it reinforces the illusion.
• 7:57
  SPEAKER_00: Another mental model appears there, confirmation bias.
• 8:01
  SPEAKER_00: This is how entire strategies get built on coincidences.
• 8:05
  SPEAKER_00: It's how we spent millions on marketing campaigns that never really mattered.
• 8:09
  SPEAKER_00: It's how technical analysis works on the stock market, or why investors load up on stocks just because they always go up in January, never realizing the data is set to January's deep, for example.
• 8:23
  SPEAKER_00: So how do we stay grounded?
• 8:26
  SPEAKER_00: That's the question.
• 8:27
  SPEAKER_00: It comes down to asking more and better questions, staying skeptical and not just taking things at face value.
• 8:37
  SPEAKER_00: What else could be driving this force?
• 8:40
  SPEAKER_00: Is there a third variable we're not seeing?
• 8:43
  SPEAKER_00: What are these second order and third order effects we talked about in part two?
• 8:48
  SPEAKER_00: Has this happened consistently or just once?
• 8:51
  SPEAKER_00: Can we explain how A logically leads to B?
• 8:55
  SPEAKER_00: Or are we just guessing here?
• 8:57
  SPEAKER_00: Or are we too eager for a neat story to please our brains?
• 9:02
  SPEAKER_00: Because sometimes there is no clean narrative.
• 9:05
  SPEAKER_00: The dog barks because of the mailman.
• 9:08
  SPEAKER_00: The bell just happens to ring at the same time.
• 9:12
  SPEAKER_00: Here's a final thought to leave you with.
• 9:14
  SPEAKER_00: When we mistake correlation for causation, we're not just wrong, we're confidently wrong.
• 9:22
  SPEAKER_00: And that's very dangerous.
• 9:24
  SPEAKER_00: Because confidence, built on coincidence, is a house of cards.
• 9:30
  SPEAKER_00: Maybe I don't know for sure.
• 9:33
  SPEAKER_00: That's how real insights begin.
• 9:36
  SPEAKER_00: So next time two things line up neatly, pause for a second.
• 9:40
  SPEAKER_00: Don't just take that at face value, don't raise the connected dots, take a moment to wonder if something else is at play.
• 9:47
  SPEAKER_00: You're in no rush, you don't have to please anyone.
• 9:50
  SPEAKER_00: Because in a world full of patterns, the best minds think independently and look for the objective truth.
• 9:59
  SPEAKER_00: That's it for today.
• 10:01
  SPEAKER_00: As always, stay curious, keep learning, and happy investing.