If you get on 538, which I do, a lot, and you get on it from your phone, your iPhone, the first thing you see at the bottom of the page is the percent chance each candidate has to become president. According to Nate Silver's model. Not the aggregated, crunched, analyzed poll numbers, which the other day were Clinton +3.5, the model's percent chance of winning, which on the same day was 70%.
I have felt all year that Silver does himself and his readers a disservice by giving his model's percent chance of winning a place of preeminence over his sophisticated poll aggregation numbers. It is very confusing and it's counter-intuitive. How can someone with a measly 3.5% advantage--And falling.--in the super-analyzed polls have a 70% chance of winning? I am not alone. Both my brother-the-trump-fuck-you-voter and my son the HRC saint asked me wtf.
Silver's answer is that the percent chance of winning is the result of simply plugging the super-charged polling numbers into his model and and hitting go 10,000 times. So 7,000 times his computer model spit out HRC having won. Okay, fine. That's his un-confusing explanation but it is still counter-intuitive. Why doesn't he display the aggregated poll numbers with the prominence he gives his model's whirs?
It is also misleading. The same day he and his gang had a podcast or a circle-jerk or something, it was the day, yesterday actually, when Silver's model showed a 70% chance of a HRC victory on a 3.5% poll advantage. This is what Silver wrote about what 70% means:
I’ve found that probabilities in the 70 percent range can be especially difficult to write about, because there’s the possibility of a misinterpretation in either direction. On the one hand, in a 70-30 race, you can usually cherry-pick your way to calling the race a tossup without that much effort, even when it really isn’t.
I have felt all year that Silver does himself and his readers a disservice by giving his model's percent chance of winning a place of preeminence over his sophisticated poll aggregation numbers. It is very confusing and it's counter-intuitive. How can someone with a measly 3.5% advantage--And falling.--in the super-analyzed polls have a 70% chance of winning? I am not alone. Both my brother-the-trump-fuck-you-voter and my son the HRC saint asked me wtf.
Silver's answer is that the percent chance of winning is the result of simply plugging the super-charged polling numbers into his model and and hitting go 10,000 times. So 7,000 times his computer model spit out HRC having won. Okay, fine. That's his un-confusing explanation but it is still counter-intuitive. Why doesn't he display the aggregated poll numbers with the prominence he gives his model's whirs?
It is also misleading. The same day he and his gang had a podcast or a circle-jerk or something, it was the day, yesterday actually, when Silver's model showed a 70% chance of a HRC victory on a 3.5% poll advantage. This is what Silver wrote about what 70% means:
I’ve found that probabilities in the 70 percent range can be especially difficult to write about, because there’s the possibility of a misinterpretation in either direction. On the one hand, in a 70-30 race, you can usually cherry-pick your way to calling the race a tossup without that much effort, even when it really isn’t.
Nobody else not named Nate Silver or FiveThirtyEight would find it difficult to write about a 70% chance of winning. No normal person would look at a 70-30 split and "cherry-pick" her "way to calling the race a TOSSUP"! This is just stoopid. 70-30 is NOT what tossup means. 50-50 is tossup, you toss the coin up in the air you have a 50% chance of heads and a 50% chance of tails.
On the other hand, a 70-30 race indicates that the bulk of the evidence lines up on one side of the case — in this case, that Clinton rather than Donald Trump will probably be elected president.
I think Silver has conjugal relations with the word "probably." He used the same term when the race, according to his model, was 80-something percent Clinton would win.
In the law and in common parlance "probably" is understood to mean 50%+1, more probable than not, that's it. "I am 80-something percent certain that is the man who robbed me," is not proof beyond a reasonable doubt, that's at least in the mid-high 90's, but it's a damn sight better than "That's probably the guy right there in the orange jumpsuit sitting next to Mr. Harris. Maybe it's Mr. Harris, but probably not." (Things that never happen in a courtroom.)
For the record, HRC's 538 lead stands at this moment, actually an hour ago on the site, at 3.1%, down from yesterday's 3.5% and continuing a steady, two-week decline. Ohio, Iowa, and Arizona are darker red than they were yesterday, and North Carolina is now light red, pink. Florida and Nevada are the palest shade of blue you can be and still be blue. Twenty-five states are some shade of blue and twenty-five some shade of red but the red is more deeply stained. The entire midsection of the country is red and all of the South except Florida. Silver projects a popular vote win for Clinton with an electoral majority of 294-243.
And, if anybody's interested, Silver pegs HRC's chances of winning at 66.5%.
Good night.
Good night.
