It still isn't. They bet on their car winning, and they lost. They didn't go around trash-talking competitors or bragging, they just thought the agreement was not financially sound for them.
So, if they had signed the agreement and won, would they have learned an $800,000 lesson in the evils of socialism or something?
They placed a bet on their car and it didn't pay off. Doesn't mean it was a bad bet. (In fact, since it sounds like they were the consensus favorites going into the competition, it was probably a good bet.)
Way to succumb to confirmation bias (EDIT: I actually meant outcome bias, thanks for the correction), article. If they won, they'd lose 600k (I'm not sure where the other 200k went, for a total of 2.5m). If they lost, they'd win 400k.
They would benefit from this only if they had less than a 40% chance to win, which, from the looks of it, they didn't. From their view, it must have looked like the other teams tried to get them (who were almost certainly going to win) to hand out some of their cash to them.
It was a rational decision, and they did make the right choice given the information available at the time. They could have been charitable, but that's not mandatory.
Confirmation bias is when you have a working theory, and only look towards evidence that supports the theory. I don't see how that applies to this article.
Oh wait, you're right. The other bias, what's it called? The one where you claim that a decision was right or wrong based on its outcome, not based on the info you had when making it?
EDIT: Outcome bias, thanks. I won't remove the error above so the discussion isn't altered.
It's fundamentally the same thing: using after-the-fact information to evaluate the likelihood of an event. Taking it one step further - using that evaluation to determine if something was "right" or "wrong" - does not change what the actual error was.
I don't see how the Prisoner's Dilemma applies here - there was no question of being sold out by the other party.
Indeed, judging by the article, they would have been very confident of being in the best 2 cars, which allows for a simple calculation of the expected return from signing the contract (the same principle applies if you want to break out the odds of coming 3rd, 4th and 5th too, but they don't change the result much - I leave the proof as an exercise for the reader ;)
if they sign the contract and come first, they win 1.7m, if they come second they win 0.4m. If they don't sign and come first, they win 2.5m, and if they come second they win 0. How confident do you need to be about coming first to not want to sign the contract? let x be the probability that you come first, we can determine what x needs to be before signing looks appealing:
x * 2.5 + (1 - x) * 0 = x * 1.7 + (1 - x) * 0.4
2.5x = 1.3x + 0.4
1.2x = 0.4
x = 0.4/1.2 = 0.333
In other words, for the contract to look attractive to them, they would have to think that they had a 33% chance or less of winning. Which considering there were only 5 cars in the race, and theirs was clearly one of the better cars, makes the contract look very unappealing.
14 comments
[ 2.9 ms ] story [ 39.6 ms ] threadThey placed a bet on their car and it didn't pay off. Doesn't mean it was a bad bet. (In fact, since it sounds like they were the consensus favorites going into the competition, it was probably a good bet.)
They would benefit from this only if they had less than a 40% chance to win, which, from the looks of it, they didn't. From their view, it must have looked like the other teams tried to get them (who were almost certainly going to win) to hand out some of their cash to them.
It was a rational decision, and they did make the right choice given the information available at the time. They could have been charitable, but that's not mandatory.
http://en.wikipedia.org/wiki/Prisoners_dilemma
EDIT: Outcome bias, thanks. I won't remove the error above so the discussion isn't altered.
And, http://lesswrong.com/lw/im/hindsight_devalues_science/
Indeed, judging by the article, they would have been very confident of being in the best 2 cars, which allows for a simple calculation of the expected return from signing the contract (the same principle applies if you want to break out the odds of coming 3rd, 4th and 5th too, but they don't change the result much - I leave the proof as an exercise for the reader ;)
if they sign the contract and come first, they win 1.7m, if they come second they win 0.4m. If they don't sign and come first, they win 2.5m, and if they come second they win 0. How confident do you need to be about coming first to not want to sign the contract? let x be the probability that you come first, we can determine what x needs to be before signing looks appealing: x * 2.5 + (1 - x) * 0 = x * 1.7 + (1 - x) * 0.4 2.5x = 1.3x + 0.4 1.2x = 0.4 x = 0.4/1.2 = 0.333
In other words, for the contract to look attractive to them, they would have to think that they had a 33% chance or less of winning. Which considering there were only 5 cars in the race, and theirs was clearly one of the better cars, makes the contract look very unappealing.