7 ms·
And yet for all their magic, they were just as wrong about the outcome of the election: You might think from a casual reading of the Cambridge Analytica press
by fovc 10y ago
And yet for all their magic, they were just as wrong about the outcome of the election:
You might think from a casual reading of the Cambridge Analytica press release that they predicted the outcome of the election. They did not. A company spokesman called reporters before election day to say that Trump had only a 20 per cent chance of winning.
Source: http://www.spectator.co.uk/2016/12/the-british-data-crunchers-who-say-they-helped-donald-trump-to-win/ http://www.spectator.co.uk/2016/12/the-british-data-cruncher...
- gareim 10y agoI'm not sure I get this logic. If I told someone they had a 16.6% chance of rolling a die and getting a 6, and then they proceed to roll the die and get a 6, would you laugh and tell me that I was wrong?
- akjainaj 10y agoI get your point but that's not a fair comparison because there were 2 choices, not 6.
- forgetsusername 10y agoI think many people would, yes. Which is frightening. I just read an anecdote about Sid Meier dealing with this in the Civilization games in Michael Lewis' Undoing Project: people couldn't stand losing battles of 33% 3 times in a row.
- pavelrub 10y agoIf it was your job to measure all the things that affect the roll of a die, and this is the result of your prediction, then I would indeed say that you failed to do your job properly. You could defend the result by claiming that it is impossible to get better measurements, and that the 16.6% chance is a reflection of our unavoidable ignorance of the system, but this doesn't really seem applicable to predicting election results. What you are essentially saying is that every prediction that doesn't claim certainty is always correct. If I "predict", by looking at a crystal ball, that there is an 80% chance of an asteroid destroying Earth tomorrow, and no asteroid destroys earth, do you think that it makes sense to say that I was correct in my prediction, because I said there was a 20% chance of it not happening? Surely you must agree that there some sense in which my prediction was wrong, or at the very least more wrong than NASA's prediction. The election predictions were wrong in the same sense.
- gareim 10y agoA lot of words to say very little. >What you are essentially saying is that every prediction that doesn't claim certainty is always correct. This seems to be the crux of it. No, what I am saying is that when you want to laugh at the polls/pollsters, you should be arguing how their methodology was wrong and what they could have done to find the true proportion +/- some standard error with however much confidence level. Simply laughing at pollsters when the minority wins is not good enough. 20% chance of winning and then actually winning does not seem all too unlikely to any reasonable person.
- pavelrub 10y ago> you should be arguing how their methodology was wrong and what they could have done to find the true proportion +/- some standard error with however much confidence level There are countless of arguments about how their methodology was wrong and what could be improved, all over the internet. Almost nobody disputes that the models were bad. I thought this part was obvious. > 20% chance of winning and then actually winning does not seem all too unlikely to any reasonable person. It seems about 80% unlikely. But the question isn't whether it seems unlikely or not, but whether it makes sense to call the prediction results "wrong". 20% chance of an asteroid not hitting Earth and then not hitting Earth might not seem extremely unlikely, but the prediction that it had an 80% chance of hitting Earth is still a bad prediction.
- delinka 10y agoThat's not a prediction of the outcome. That's a statistic. A prediction of the outcome would be "you won't roll a 6."
- Jarwain 10y agoThe article doesn't claim that Cambridge Analytica predicted the outcome. The only factual claim about the outcome made by the company, according to the article, is that statistic. The article, however, took that statistic and implicated it as a prediction when all it really was was a statistic.
- dagw 10y agoIf you asked me to predict whether you will roll a six or not, then I will predict that you won't. Is that a bad prediction?
- delinka 10y agoIt's certainly an incorrect prediction if I roll a six. You didn't simply state the odds and leave it at that, you actually made a prediction. Is it a "bad" prediction? Depends on how you're defining "good" vs. "bad" predictions. If the quality of the prediction is not based on the outcome (i.e. whether the prediction is correct), but is instead based on the odds themselves, then that's a conversation that leads to "bad but correct" predictions.
- arrel 10y agoThe difference is that the roll of a die is random, but the outcome of an election gets more deterministic as it approaches. Election prediction is figuring out what people have already decided they are going to do.
- BigJeffeRonaldo 10y agoWell, the model at least does not maximize likelihood
- forgetsusername 10y agoAnalysis is useless if it isn't right each and very time, got it.
- endless1234 10y agoHow does Trump winning prove he had more than a 20% chance of winning?
- delinka 10y agoI feel this is the rub with this particular comment thread. If all they gave were stats, then they didn't make a prediction.