What in the world is Nate Silver talking about. He says he would bet on Romney. WTF?

[Hamlette]

"So I would bet on Mr. Romney right now given the odds the model offers — but I’d have done so more confidently before the morning’s jobs report.

It’s going to take a few more days for the forecast model to catch up to the news, and I don’t think there’s any alternative but to keep an open mind about the polls for right now."

He's talking about the Rassmussen poll and the We Ask America polls which show a big jump for Romney in Ohio, VA and FL. He still has Obama at a 84% chance of winning in his forecast (I can't access his "now cast" on my computer...not sure why, been like that for weeks)

So Nate would bet Romney would win? What am I missing?

http://fivethirtyeight.blogs.nytimes.com/2012/10/05/oct-5-day-after-debate-strong-swing-state-polls-for-romney/

[NRaleighLiberal]

1. I read that and think it is poorly written - he meant bet on Romney's debate impact to raise him

a bit in the polls...but it was sloppy grammar.

my thought, anyway....

[Kalidurga]

2. I think he means the pay off would be much larger...

I still wouldn't bet on Romney though, even if it was a sure thing and even with a bigger pay off.

[budkin]

3. He's saying he bets the shifts came from Romney's strong debate, not randomly

However, it does not have the advantage of knowing that the shifts may have come for a good reason — in this case, Mr. Romney’s strong performance in the debate.

[7worldtrade]

4. Betting on a Romney bounce

In the context of Nate Silver's posting, I think he meant he bets that the movement in the polls is in fact a bounce (even if temporary) for Romney rather than statistical noise, but he's not sure and needs to see more polls, although he's less confident there will be a bounce in light of the jobs report. He just doesn't know. At least that's how I read it.

[Hugin]

5. If you read a paragraph or so up from there, 528 is saying...

Romney went from like a 12% to 15% chance of winning. What you posted seems to be a discussion of if this increase is real or not and 528 is saying that he feels that the increase is real. Not necessarily that Romney will win. Which, I tend to agree with given his odds are still only 15% after the first debate.

[DemocratSinceBirth]

6. He's Saying Romney Is A 5-1 Dog

5-1 underdogs win about one out of every 6 times.

[Hamlette]

7. I hope you are all right. Shocked me. Couldn't see any comments so didn't leave one on the blog.

[hlthe2b]

8. I think he is saying he is betting on Romney's apparent bounce being real... not that he'd win

[Cicada]

9. Silver is only saying Romney has greater than 15% chance to win

The polls caused his model to change from Romney having 12% chance to win to 15% chance to win. Silver thinks Romney has some chance greater than 15%. In a bet his model would pay a winning Romney better $100 for a bet ticket costing only $15. If the true odds were 17%, say, the correct price for the ticket would be $17. Silver merely says he would pay - bet - $15 because he thinks the ticket is really worth - say - $17. Play the betting game 100 times and you would win $1700 - winning $100 17% of the time. But his model would only charge you $1500 for playing 100 times. You would make a $200 profit over the 100 bets.

[railsback]

10. Seems Nate is totally disregarding the after effects of Romney's debate lies

Seriously, that is SOMETHING you have to take into account. Sure, Romney was blasting away, but with what? Total shit. It seems Nate is just going off the initial reaction, not the true results based on the countless evaluations by fact checkers. You gave the conventions a full week plus to analyze the results, why treat the debates any different? Makes NO sense whatsoever.

[boxman15]

11. He's not saying he thinks Romney will win.

Because of those odds (roughly a 4 in 5 chance of an Obama win), he believes it would be smart to put money on Romney to win. It's not impossible for him to pull it off and if you bet, Silver thinks it would be wise to put money on Romney.

[muriel_volestrangler]

12. Bet on Romney at the odds you can get now; Silver thinks the payout will decrease

so that by betting now, you get the largest payout available between now and the election. But the Obama payout will increase at the same time, so, if the shift is enough, you will be able to bet on Obama as well, some days from now, and guarantee a profit whoever wins.

eg, on a British betting site,
Romney is at 11/4 now (meaning bet $4, you get that back, plus $11), Obama at 1/4
This translates to the odds of Romney winning being reckoned as ( 4/(4+11) )/( 4/(4+11) + 4/(4+1) ) = 0.25

If Romney improves to 9/5, Obama would go to about 2/5
This translates to the odds of Romney winning of ( 5/(5+9) )/( 5/(5+9) + 5/(5+2) ) = 0.33

(these odds allows the bookmaker to make the rough profit they expect, if everyone where to bet at those odds - 5/(5+9) + 5/(5+2) = 1.071, which should make them a profit of 7.1% of the bets laid, if they've gauged the odds right; the equivalent figure for the earlier odds is 4/(4+11) + 4/(4+1) = 1.067 )

If you bet $28 now on Romney at 11/4, you'd get back, if he won, (28/4) * (11+4) = $105
If you bet, in the future, $75 on Obama at 2/5, you'd get back, if he won, (75/5) * (5+2) = $105

So, you've have bet a total of 28+75 = $103, but you'd get back $105, whoever won. Since you correctly guessed the bookmakers would change the odds to say Romney is more likely to win than now, you have made a profit.

[grantcart]

13. he is talking about betting on Romney getting a temporary bubble out of the debate.