Reply With Quote
^Meant to post this under my real avatar.
http://www.slate.com/articles/news_a...al_genius.html
Very good article. Points out why his fans as well as his critics (like me) are wrong. But I've already admitted that I was wrong about the polling data.
It's well after midnight on the East Coast, and the results are in: Nate Silver has won the 2012 presidential election by a landslide. His magic formula for predictions, much maligned in some corners in recent weeks, appears to have hit the mark in every state—a perfect 50 green M&Ms for accuracy. Now my Twitter feed is blowing up with announcements of his coronation as the Emperor of Math and the ruler of the punditocracy. Wait—it was even more than that, they say: a victory for blogging, and also one for rational thought. He proved the haters wrong! He proved science right! Is this guy getting lucky tonight or what?
But all these stats triumphalists have it wrong. Nate Silver didn't nail it; the pollsters did. The vaunted Silver "picks"—the ones that scored a perfect record on Election Day—were derived from averaged state-wide data. According to the final tallies from FiveThirtyEight, Obama led by 1.3 points in Virginia, 3.6 in Ohio, 3.6 in Nevada, and 1.9 in Colorado. He won all those states, just like he won every other state in which he'd led in averaged, state-wide polls. That doesn't mean that Silver's magic model works. It means that polling works, assuming that its methodology is sound, and that it's done repeatedly.
To be fair, the art of averaging isn't simple. You can't just comb through Google News and add up all the numbers that you find; to find a useful sum, Silver judges which polls to put into his analysis and then he weighs them according to his perception of their quality. He may be very good at this, but other stat-head pundits do more or less the same, and their averages match up accordingly. Yes, Silver had Obama in the lead, but so did RealClearPolitics and Talking Points Memo; where he had Romney, so did they. "Our state-by-state forecasts are extremely similar to those issued by our compe ors," he wrote in a post two weeks ago, en led "State Poll Averages Usually Call Election Right." Those numbers nailed it on 50 out of 50 states, as they often do.
So picking winners state by state was the easy part. Anyone who glossed the numbers would have made the same projections. But Silver's model promised more than that: He offered assessments of his confidence in each state's results. The fact that Obama led in Ohio polls made it obvious that he should be the favorite, but what if those Ohio polls were wrong? How much risk was there in trusting state-wide averages? This was Silver's nifty contribution: He assigned that risk a probability, by looking at some other factors, such as polling trends and local demographics. Take the example of Virginia, where Obama led by 1.3 percentage points. Picking him to win the state was a no-brainer since he was leading in the polls, but Silver used his secret sauce to calculate the chances that those polls were wrong. According to his calculations, the risk was 21 percent, meaning that Obama's odds to win the state were roughly 4-to-1.
What do the day's returns tell us about the accuracy of Silver's model? Nothing much. The fact that Obama won Virginia looks good for averaged polling—indeed, his margin appears to be a couple points, not far off from what was predicted—but we'll never know about that other part. Did Obama really have a 79 percent chance of winning? To get a sense of that, we'd need to run yesterday's election like a lab experiment, doing it 10,000 times to see how often Obama wins. Since that can't happen, we're left to scratch our heads.
You could even make a case that Silver's estimates were off. His averages had Republican Senate candidate Rick Berg up by 3.9 in North Dakota, enough to give him a 92 percent chance of winning. Now it looks like he could lose to Democrat Heidi Heitkamp. In Nevada, Silver gave Republican Dean er an 83 percent chance of winning based on a 4.7-point lead in the averaged polls. As of now, er’s up by just 12,000 votes.
But Montana is the most telling case: According to Silver's polling average, the Democratic candidate, Jon Tester, led by 1.4. But Silver's model, which uses “state fundamentals” among other factors, guessed the polls were wrong, and gave his opponent Denny Rehberg a 66 percent chance of winning. So far as I can tell, this was the only contest in which the magic model went against the averaged polls. Guess what? The projections might have been a little off: Tester has a 5-point lead.
It’s possible that Silver’s predictions were weaker than normal in North Dakota and Montana because polling in those states was sparse. Again, the polls are more important than the poll watcher—good polling yields good predictions. In the future, I'll be curious to see how Silver's model does in cases like Montana, where it picks the polling underdog. Does his secret sauce yield some unexpected scores—winners who surprise the pundits and the pollsters—or does it just distract us from the obvious?
Silver lovers aren't waiting for these comparisons. They're riding high on victory, and giving credit to the bearer of good news. In doing so, they’ve made the same mistake that Silver's critics made last week: They've confused his projected odds with hard-and-fast predictions, and underestimated the accuracy of polling. The fact that Obama won doesn't make Nate Silver right, any more than a Romney win would have made him wrong.
^Meant to post this under my real avatar.
why people need two accounts is beyond me.
Created that in 2008 to troll AngelLuvs' classic thread.
Those fools Rove and Morris on FoxNews sure thought they had it right.
morris blamed the polls on media bias and polling democrats.
turns out his predictions were nothing more than media bias.
Genius? I don't know
Good statistician that has proven himself over the course of 3 general elections and many smaller ones? Absolutely!
There's some seirous butthurt through the media at all the love Nate Silver has gotten. Many have tried to write off his predictions and their accuracy but the one thing that none of them can answer is this:
If what Nate Silver is doing is so easy why aren't there more people doing it just as well?
?There are obviously a ton of people out there trying to but they're not able to duplicate his results. Why?
Your conformation bias is amazing, Darrin. A day after you might have gained a bit of humility after how wrong you were on this front and you're still trying to find ways to show that Nate Silver is wrong and you are right. One day you might learn to start trusting the data even when its telling you things you don't want to hear.
I would actually argue that's a pretty poor article. Describing Silver's method as mere "poll averaging and weighting" is, well, not really correct. As pointed out many times, there are other factors that are included in his model, such as economic data, etc.
Also, the author misinterprets what the "chances of winning" means. Silver *does* run thousands of simulations on his model and arrives at those predictive percentages. That the actual results (there were predictions for many races) closely matches such output, it's an actual validation for the model (at least it was yesterday).
For contrast, you can take the vaunted (by some) University of Colorado model, that allegedly was modeled to predict correctly elections since 1980, and after predicting a 300+ electoral vote victory for Romney, one would be hard pressed not to call such model flawed.
lol the level of butthurt here is epic
Either candidate could have won and you would not be able to say anything about Silver's model (the odds). You'd have to run the election numerous times, but you can't.
Nothing really magical about what he is doing and I'm not criticising his methods? I didn't trust the DATA -- I thought the polls oversampled dems and I was WRONG about that.
Here's the magic of a Nate Silver-esque model on a web page.
http://www.270towin.com/simulation/
Now, if you were to click the "Run Again" button some number of times, n, and divide the number of Obama simulated wins by n, you'd have Obama's odds of winning. It's no more complicated than that.
because fox told you to think that.
Are Nate Silver's odds exactly correct? Who ing knows? Is it a relevant question? No. The important thing to take away is that Nate Silver's modeling is a very good way to take polling data and give you a very good indicator of how a state will go. Its not just that he picked the appropriate winner in 98 out of 100 state elections in 2 consecutive presidential elections but the fact that he's been able to point out to which battleground states are actually compe ive and to what degree. His model had Florida as the closest battleground state and guess what - it was. The devil is in the details but the details are exactly what you're not interested in, Darrin.
I asked you what problem you had with the model and you responded with the polling data itself. But why? There was basically no reason for you to be skeptical of polling data because it has historically been shown to be a fantastic predictor of elections. You had nothing to go off of. If you can provide a data based reasoning for how you come to your conclusions I would love to hear it.
There's no magic in his secret sauce. The polls accurately predicted the election.
EDIT> His "modeling" is something anyone with a VERY BASIC understanding of statistics could do themselves.
LOL who the has said anything about it being magical? Its ing statistics. What was your reasoning for not trusting the data outside of some red team pundits telling you what you should have thought? On the other hand, Nate Silver has years of blog posts detailing why he trusts polling data where he gives expliciting reasons and data to back up said reasons.
No Darrin, its not magic. Its just logic. Which might as well be magic as far as your thought process is concerned.
Thanks for explaining to us that its not magical, Darrin. I'm sure we were all under the impression Nate Silver was walking around his cauldron casting spells and doing his best Harry Potter impersonation.
Now, can you explain why you thought the poll data was wrong? Can you provide a historical background showing such concerns to be valid?
2012 was the first year polls accurately predicted the election. Except not.
Silver is a boss. Too bad it sounds like he might be moving on from politics. Then again, hopefully he can use his skills elsewhere and have similar success.
The odds are eye candy. The meat is on the predictions, and that's where everybody (pollster or aggregation) eventually gets judged. You can't fake the predictions, and you don't need to run multiple election to know how off or not the model is.
Not to mention there were really awful polls out there in the last couple days (see Rasmussen), which really makes the argument that all he does is average polls just a complete fallacy.
Where's the economic data there? How would you weight such data? How are the polls weighted?
It's all easy as pie until you actually go and make what you think it's a reliable model then it blows in your face. Ask the Unskewed Polls guy or Scott Rasmussen.
What Silver does is not secret sauce simply because he has published here and there what his methodology is. I expect other pollster to draw on that going forward.
It's amazing how much Nate Silver owns Darrin.
Yet, you doubted him. Hmm . . .
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
