It’s a benefit to be hardwired to see the tiger (signal) in the grass (noise) because there’s little penalty if there isn’t a tiger and a huge one if there is and you miss it.
This is, IMO, why AGW is so popular. Weather is fractal, if it is suggested that danger exists from a given set of weather patterns, people will begin to see those patterns – even if the actual pattern (if there is one) is not really in the noise because it’s time constant is too long. The problem there is that the penalty for the no tiger case is roughly similar to the tiger case.
Which also explains why so many hysterics and alarmists in AGW land.
There are many agents at work. People who are related tend to live near each other. Old people tend to retire in the same areas. Eating, smoking and exercise habits tend to be similar region to region. And, after all, one in three people will develop cancer in their lifetime.
With this paragraph the writer undermines his own argument by saying there is a cause, if hidden. We very often do see patterns that don’t exist and it is a powerful delusion in many cases. I’ve written that the gambler’s fallacy, which is related, is almost universal… by which I mean, universal!
Many would like to think they are not superstitious, but we all see patterns that are not as meaningful as we make them. Being human sucks. Other options suck more.
One funny thing about the gambler’s fallacy is that the opposite may actually be true. If a roulette wheel has had ten red in a row the fallacy says black is due. However, more likely is there are defects in the wheel that make red more likely. Place yer bets.
Without taking anything away from the sharpshooter fallacy; If you see a barn with just one very tight cluster, perhaps with a stray hole here or there, would it be wrong to assume someone was firing at that one spot?
The discussion on the novel titled “Futility” and Titanic beings to mind Buzz Aldrin’s 1995 novel, “Encounter with Tibor”. It starts with a Shuttle accident in late 2002 that results in a commission investigating the accident to recommend NASA should stop going in circles and start exploring the Solar System again, the Shuttle is not followed by a RLV but instead is replaced by the “Pigeon”, a capsule modeled after the Apollo which is used to return astronauts to the moon.
Perhaps future generations will look for connections between it and NASA space policy in the last few years 🙂
For the polar opposite of this article, see Arthur Koestler’s amazing book “The Roots of Coincidence.” A highly readable tome by a first-rate intellect, it will have the gullible convinced (not just believing: convinced) that everything is tied together by quantum mechanics.
An interesting and well-written article. I only wish the author had clarified that we have rational mathematical tools that enable us to distinguish signal from random noise. In fact, as an occasional statistical consultant, it seems clear to me that the only way to distinguish patterns from chaos in the data we collect is through the use of statistics. So when he quotes the New Yorker saying that no cancer clusters have ever been shown to have a convincing link to an environmental cause, he is referring to statistical hypothesis testing. Without this insight, the article tends to slouch into the “can’t trust anything, there are no real patterns” posture. This is bad because there are real patterns out there. Smoking cigarettes really does increase your chance of getting cancer or heart disease; that’s not just confirmation bias.
His one attempt at mathematical calculation is a bit off, too. The chance of drawing one particular sequence of 10 cards from a deck of 52 is 42!/52! or about 1 in 57 quadrillion, not trillion.
BTW, I’ve never heard it called “The Texas Sharpshooter Fallacy”, only “The Sharpshooter Fallacy”. As a quondam Texan I’m mildly offended 😉
I’ve been playing blackjack for 20 years, and one thing I’ve noticed is how many “impossible” streaks regularly occur. One particular card sequence occurred so regularly for so long, that I started to “predict” the outcome (my loss) to impress the dealer. After what seemed a few months, the unusually frequent occurrence disappeared. The patterns we “see” in random events can sometimes seem amazing, and of course that’s a big reason casinos stay in business. One event like this caused the death of a close friend’s uncle. He spent several days on an amazing lucky streak at the roulette wheel. He then decided that he had “figured out” how to win, so he got his life savings and proceeded to lose it, and then went home and put a bullet through his head.
My dad used to take me to Vegas when I was a teenager and we both played roulette and blackjack. When I got older I would write a simulation to test any betting scheme I’d come up with. In the simulation, they all lost no matter how much money I’d start them with or what limits I’d allow for the game. No matter how you bet, there will always be a string of losses long enough to take whatever amount you bring. There will also be a string of wins that a ‘let it ride’ would give you a huge win… but you never know when it will happen and after a big enough string of losses, you don’t get the chance to find out.
Life is a gamble, but putting it all on the line is not usually a good idea. Double or nothing guarantees nothing if you play long enough.
Ken, we have a state lottery out here, and since it’s a monopoly, the odds are horrible. Anytime someone informs me that they “play the lottery” I offer to stand in lieu of the state — it’s simple enough to simulate as many plays as you want on a spreadsheet. I then ask “how many times do you want to play?” and generate as many “tickets.” The result is always the same: I announce, “Your numbers didn’t come up — you lost, now hand over the money.” It’s blunt and humorous, but I think the point is driven home every time.
It’s a benefit to be hardwired to see the tiger (signal) in the grass (noise) because there’s little penalty if there isn’t a tiger and a huge one if there is and you miss it.
This is, IMO, why AGW is so popular. Weather is fractal, if it is suggested that danger exists from a given set of weather patterns, people will begin to see those patterns – even if the actual pattern (if there is one) is not really in the noise because it’s time constant is too long. The problem there is that the penalty for the no tiger case is roughly similar to the tiger case.
Which also explains why so many hysterics and alarmists in AGW land.
There are many agents at work. People who are related tend to live near each other. Old people tend to retire in the same areas. Eating, smoking and exercise habits tend to be similar region to region. And, after all, one in three people will develop cancer in their lifetime.
With this paragraph the writer undermines his own argument by saying there is a cause, if hidden. We very often do see patterns that don’t exist and it is a powerful delusion in many cases. I’ve written that the gambler’s fallacy, which is related, is almost universal… by which I mean, universal!
Many would like to think they are not superstitious, but we all see patterns that are not as meaningful as we make them. Being human sucks. Other options suck more.
One funny thing about the gambler’s fallacy is that the opposite may actually be true. If a roulette wheel has had ten red in a row the fallacy says black is due. However, more likely is there are defects in the wheel that make red more likely. Place yer bets.
Without taking anything away from the sharpshooter fallacy; If you see a barn with just one very tight cluster, perhaps with a stray hole here or there, would it be wrong to assume someone was firing at that one spot?
The discussion on the novel titled “Futility” and Titanic beings to mind Buzz Aldrin’s 1995 novel, “Encounter with Tibor”. It starts with a Shuttle accident in late 2002 that results in a commission investigating the accident to recommend NASA should stop going in circles and start exploring the Solar System again, the Shuttle is not followed by a RLV but instead is replaced by the “Pigeon”, a capsule modeled after the Apollo which is used to return astronauts to the moon.
Perhaps future generations will look for connections between it and NASA space policy in the last few years 🙂
For the polar opposite of this article, see Arthur Koestler’s amazing book “The Roots of Coincidence.” A highly readable tome by a first-rate intellect, it will have the gullible convinced (not just believing: convinced) that everything is tied together by quantum mechanics.
An interesting and well-written article. I only wish the author had clarified that we have rational mathematical tools that enable us to distinguish signal from random noise. In fact, as an occasional statistical consultant, it seems clear to me that the only way to distinguish patterns from chaos in the data we collect is through the use of statistics. So when he quotes the New Yorker saying that no cancer clusters have ever been shown to have a convincing link to an environmental cause, he is referring to statistical hypothesis testing. Without this insight, the article tends to slouch into the “can’t trust anything, there are no real patterns” posture. This is bad because there are real patterns out there. Smoking cigarettes really does increase your chance of getting cancer or heart disease; that’s not just confirmation bias.
His one attempt at mathematical calculation is a bit off, too. The chance of drawing one particular sequence of 10 cards from a deck of 52 is 42!/52! or about 1 in 57 quadrillion, not trillion.
BTW, I’ve never heard it called “The Texas Sharpshooter Fallacy”, only “The Sharpshooter Fallacy”. As a quondam Texan I’m mildly offended 😉
I’ve been playing blackjack for 20 years, and one thing I’ve noticed is how many “impossible” streaks regularly occur. One particular card sequence occurred so regularly for so long, that I started to “predict” the outcome (my loss) to impress the dealer. After what seemed a few months, the unusually frequent occurrence disappeared. The patterns we “see” in random events can sometimes seem amazing, and of course that’s a big reason casinos stay in business. One event like this caused the death of a close friend’s uncle. He spent several days on an amazing lucky streak at the roulette wheel. He then decided that he had “figured out” how to win, so he got his life savings and proceeded to lose it, and then went home and put a bullet through his head.
My dad used to take me to Vegas when I was a teenager and we both played roulette and blackjack. When I got older I would write a simulation to test any betting scheme I’d come up with. In the simulation, they all lost no matter how much money I’d start them with or what limits I’d allow for the game. No matter how you bet, there will always be a string of losses long enough to take whatever amount you bring. There will also be a string of wins that a ‘let it ride’ would give you a huge win… but you never know when it will happen and after a big enough string of losses, you don’t get the chance to find out.
Life is a gamble, but putting it all on the line is not usually a good idea. Double or nothing guarantees nothing if you play long enough.
Ken, we have a state lottery out here, and since it’s a monopoly, the odds are horrible. Anytime someone informs me that they “play the lottery” I offer to stand in lieu of the state — it’s simple enough to simulate as many plays as you want on a spreadsheet. I then ask “how many times do you want to play?” and generate as many “tickets.” The result is always the same: I announce, “Your numbers didn’t come up — you lost, now hand over the money.” It’s blunt and humorous, but I think the point is driven home every time.