| |
Yes -- polls were showing her ahead.
Well, as you'll recall, Trump's fellow Republican, James Comey, who had earlier tried to destroy Clinton's campaign by using a big press event as an opportunity to trash her personally, even as he admitted there was no basis for indictment, intervened in the election again, with just days to go, by speculating, entirely baselessly, that some emails on a former Congress member's laptop might contain something incriminating Clinton. That gave the corporate media an excuse to spend the final week of the campaign doing negative coverage for Clinton, depressing her poll numbers several points, putting the election close enough that Trump, with the benefit of the effective "home field advantage" Republicans get in the Electoral College, could manage a minority win.How did she manage to snatch defeat from the jaws of victory?
Was it?It's one of the biggest upsets in election history.
What makes you think that? I don't recall anyone saying anything of the sort. Certainly I can't imagine anyone who actually followed Silver's work, which mostly is in the field of sports statistics, would have said that, since by nature the team he gives the lower probability of winning a game is going to be victorious a significant minority of the time.But, Nate Silver is NEVER wrong!
Nate Silver's final forecast gave Hillary Clinton a 71.4% chance of victory. Was that wrong? Well, in one sense, any prediction he might make other than saying that one or the other had a 100% chance of victory would be wrong, since ultimately one candidate or another was going to 100% win. If Clinton had won, then in the same sense he'd have been wrong to say Trump had a 28.6% chance, when in fact it was 0. But that's not a very sensible way to think about probabilities.
Instead, think of it the way you would if this were a sporting event. Let's say an announcer told you that an NFL kicker, lined up for a 48-yard-long field goal, had a 71.4% chance of hitting it.... and then the kicker missed. Would you say the announcer got it wrong? What if, in fact, NFL kickers do tend to hit 71.4% of their kicks from that distance? Wouldn't that, then, have been a pretty accurate estimate of the odds, even if that particular kicker ended hitting 100% or 0% of that particular kick?
Who are you quoting? I Googled "crush his nuts" and "Hillary Clinton" and the first Google result was to that post you just made, so it does not seem to have been a common prediction at the time.She was going to "crush his nuts"!
In my case, once, just now, from you. How about you?How many times did we hear those?
Last edited by Arkady; 13th June 2018 at 12:19 PM.
Now, tell me how you'd react if we simply flip the politics around on this. Let's say, for example, that Trump were to justify his travel ban policy by pointing out that Muslim terrorists killed three thousand people on 9/11. Would you say Trump was lying because, in fact, it was a little less than three thousand?
No need to answer that. We both know you wouldn't. In fact, we both know that if a liberal were to be so absurdly petty as to call Trump a liar on the basis of that, you'd rightly point out the stupidity of that attack. You'd see, in that case, what a brain-dead attempt at political "gotcha" it was. But that's what partisan politics can do to a person.... the same attack he'd scoff at if the politics were arranged one way, he'll eagerly engage in when they're arranged the other way.
Last edited by Arkady; 13th June 2018 at 12:20 PM.
By what standards? I would have thought the biggest upsets to be ones where expectations were for a landslide in one direction and things actually went the other way. In the case of the Clinton/Trump election, I didn't know anyone who expected a landslide. There hadn't been a true landslide in many cycles, and the final polling average only had Clinton up by 3.2 points, which is about the size of the margin of error of polls. I expected a very close contest and that's exactly what we got, with Clinton beating Trump by about three million votes in the popular vote, while Trump took the electoral college by way of a fraction of a percentage lead in three key states.
I wasn't around, so I can't say. But I certainly never saw it until just today, from you."Crush his nuts" was said many, many times on this forum.
Since Silver has worked for many years doing sports stats, try to think of it in those terms. Let's say Silver claims that Lebron James has a 70% chance of making a free throw. If you watch James shoot ten free throw shots and he misses three, will you say Silver was wrong three times? I wouldn't, because that's not how probabilities work. In fact, if James missed three shots of out ten, that would suggest Silver was right about his probability of hitting each shot.We also heard over and over about how Nate Silver had never been wrong.
In the same sense, if Silver makes ten different predictions of a 70% chance of a politician winning an election, and three of those times the politician he picked loses, does that mean he was wrong three times? I wouldn't say so, since, again, that's not how probabilities work. Rather, we'd fully expect three of the favored candidate to lose, if Silver was right about the chances. If, instead, the candidate he said had the better chance won every single time, THEN we'd say Silver was wrong, because then he would be underestimating the chances of victory.
Now, I haven't gone back to see how accurate Silver is overall, by checking the percentage he ascribes to all candidates and seeing how the actual elections tend to turn out. Does he systematically overestimate or underestimate the probable winner's probability? No clue. But, obviously, taking a one-election snapshot and arguing it shows Silver was wrong, when he states his predictions in probability figures, makes no more sense than saying he's wrong about a 70% free-throw shooter if we see the guy miss.