September 20, 2006


Polls, Pundits Tout GOP Gloom, But Smart Money Bets Different (JED GRAHAM, 9/19/06, INVESTOR'S BUSINESS DAILY)

[T]hose who put money behind their predictions see things differently. Quotes at show that futures traders favor Republicans to retain control of the House.

Late Tuesday, TradeSports' futures contract for GOP House control was trading at 53.8 — a 53.8% chance. TradeSports gives an 80.3% chance that Republicans will hold onto the Senate.

Betting exchanges, or futures markets, are worth paying attention to because they have a track record that would put most pundits to shame.

In 2004 the Iowa Electronic Markets, a research vehicle operated by the University of Iowa, gave President Bush a 51.4% chance of winning re-election vs. a 48.6% chance for a Sen. John Kerry victory. Those totals were eerily close to the popular vote results.

The prediction markets tend to do better than polls because they are able to capture "the wisdom of crowds" and consider questions a pollster wouldn't think to ask, said Don Luskin, chief investment officer at TrendMacrolytics.

Of course, the folks predicting a Democrat landslide have all their money tied up in oil futures.....

Will the Democrats Flip the House?: Slate's mathematician on the odds of a Democratic victory. (Jordan Ellenberg, Sept. 19, 2006, Slate)

The simplest model is to assume that the Democrats have a 50/50 chance of winning each race and that the races are mutually independent. That is, if we knew that the Democrat won in North Carolina, it's no more or less likely that the Democrat won in New Mexico. (You can find a more thorough discussion of independence, not to mention a stiff dose of 2001 nostalgia, in my article on Gary Condit's love affairs.) The theorem you need to know here is the Law of Large Numbers: If you have a bunch of independent 50/50 chances, it's very likely that just about half of them will go one way and half the other. In other words: Not only can you not expect everything to go your way, you can't even expect most things to go your way. Chances are that Democrats will win very close to half of the 24 tossups. Half is 12, and 12 is not enough.

The Law of Large Numbers suggests that it's unlikely Democrats will win as many as 15 of the 24 tossups. To guess exactly how unlikely requires a more powerful tool: the Central Limit Theorem. Imagine sketching a graph of the probability of various outcomes of the midterm election. The graph would have a big hump in the center, representing the most likely scenarios—the Democrats winning around 12 tossups—and would rapidly tail off in both directions as the outcomes became more extreme. What you just drew is the normal distribution, or bell curve—the foundation of all applied statistics. The Central Limit Theorem says that a random variable (like the midterm election) made of many small, identical, independent parts (like the individual House races) always obeys a law approximating the bell-shaped curve. The more individual events involved, the better the obedience to the bell curve.

If you flip a coin—or a congressional district—N times, you expect to get about N/2 heads. But it's too much to expect to hit N/2 on the nose—it's standard to see some deviation from dead center. But how much? The answer, naturally, is called the standard deviation, which in this case comes to half the square root of N. So, for our 24 tossup elections, the standard deviation is about 2.45. We shouldn't be too surprised, then, to see between nine and 15 seats go to the Dems. But the Central Limit Theorem does this one better; it says that the chance of beating expectations by one standard deviation or more is approximately 16 percent. To win control, the Democrats have to beat expectations by three seats, a bit more than one standard deviation; their chance of doing so should thus be a bit less than 16 percent. In fact, it's about 15.4 percent.

G.O.P. Gains Big Fund-Raising Advantage (JEFF ZELENY, 9/21/06, NY Times)
House Democrats have nearly as much as Republicans, while Senate Democrats have $10 million more than their Republican counterparts.

But that edge is being offset by the fund-raising of the Republican National Committee, which has said it would spend more than $60 million to make up for any shortcoming by the party’s Congressional committees.

The cash-on-hand figures, revealed in reports filed with the Federal Election Commission, are critical indicators of the resources the parties can draw upon during the intense days ahead. The money is primarily devoted to television advertisements in districts that are — or will become — competitive, as well as to help finance get-out-the-vote operations.

The new fund-raising reports show that at the beginning of September, Republicans had $39 million in the bank, compared with $11 million for Democrats.

Posted by Orrin Judd at September 20, 2006 9:39 PM

Just a few weeks ago -- when the polls were at their worst for the GOP -- GOP control had dropped to a low of around 46 on tradesports.

Elections aren't held in August. You'd think the Dems would have learned that by now.

Posted by: kevin whited at September 20, 2006 10:36 PM

Oil futures? How about gold?

Posted by: ratbert at September 21, 2006 12:22 AM

After today, they've probably spent what cash they have on winter vacation packages in Caracas.

Posted by: John at September 21, 2006 12:37 AM

I've got a $10 a seat swing bet with one of my liberal Democrat friends who is sure that they'll take over both houses. If I'd made the bet a month earlier he'd have likely given me odds. He's not as sure now as he was.

Posted by: Rick T at September 21, 2006 5:57 AM

If there are any smart Democrats left, they would invest their money in cattle futures. That's where the big money is. Right?

Posted by: AllenS at September 21, 2006 6:51 AM


Only the real rightwing whackos are in gold.

Posted by: oj at September 21, 2006 8:39 AM

The only problem with the mathematician's analysis is the Central Limit Theorem and Law of Large Numbers requires that the seat results are independent. That's not clear; if the "Toss Up" status is based on nationwide polling rather than local data, any error in the polling affects all the races simultaneously.

Posted by: John Thacker at September 21, 2006 10:03 AM

And Slate's mathematician apparently sees no difficulty in applying the Law of Large Numbers when n is as small as ... 24?

Posted by: Anthony Perez-Miller at September 21, 2006 11:20 AM

Hey, cut the guy some slack. It's hard pursuading the "Reallity Based" "Truther" "community" with facts, especially numbers and statistics, and at least this guy is trying. (Not that it'll do any good.)

Posted by: Raoul Ortega at September 21, 2006 11:32 AM

There are more problems than John hints with the Slate "mathematician"'s analysis. With only 24 cases the distribution is Binomial, not Gaussian - it becomes Gaussian with a large number of cases (the point of the Central Limit Theorem), but this will never happen in this case - hence any mention of the Central Limit Theorem is totally irrelevant and is apparently included only to give increased credibility. Fortunately, using the handy Binomial Calculator the odds can be calculated precisely: to achieve at least 15 seats out of 24, assuming a 50% chance for each seat, has a probability of 15.37%.

Of course, this simple model is unreasonable; the contests are not totally independent. A better guess would be the Delphic Tradesports pool which is currently at 43.5% for a Dem takeover, or Election Projection 2006 which is currently predicting only an 11 seat gain for the Dems.

Posted by: jd watson at September 21, 2006 11:33 AM