# Outside the Beltway

## Fun with Stats

Steve Bainbridge is going to be pulling extra hard for his Redskins after reading this odd statistic in the LA Times [otbblog/jamesotb]:

A 72-year streak links the victory or defeat of the Washington Redskins on the eve of election day with the presidential race. If the Redskins go down to defeat or tie, the sitting president’s party loses the White House.

***

The Redskins’ performance has aligned with the presidential outcome in the last 18 elections Ã¢€” a probability of 1 in 263.5 million, according to Dave Dolan, an assistant professor of statistics at the University of Wisconsin-Green Bay.

A rather inprobable coincidence, to be sure.

FILED UNDER: Sports
James Joyner is Professor and Department Head of Security Studies at Marine Corps University's Command and Staff College and a nonresident senior fellow at the Scowcroft Center for Strategy and Security at the Atlantic Council. He's a former Army officer and Desert Storm vet. Views expressed here are his own. Follow James on Twitter @DrJJoyner.

1. Wow.

Errr, go Redskins! Unless they are playing the Steelers. Then . . . well, I’ll just say how this is a stupid statistic.

2. (Sorry for the double post.)

Who has the time to come up with these things, anyway?

3. Tom N says:

This is going to kill Oliver Willis.

4. Paul says:

The RedskinsÃ¢Â€Â™ performance has aligned with the presidential outcome in the last 18 elections Ã¢Â€Â” a probability of 1 in 263.5 million, according to Dave Dolan, an assistant professor of statistics at the University of Wisconsin-Green Bay.

HUH? I ain’t done the math but my B.S. detector just went off on that number. (although one would assume the guy is right)

5. Paul says:

or Maybe not… I thought of it as a coin flip but is is a Pascal’s triangle problem

6. James Joyner says:

I’m also not sure if he’s only considering the two major parties, all known candidates, or what.

7. jen says:

Hail to the Redskins.

The good news is that Gibbs is back and by November the team should be in good form under his leadership. That is, if you’re for President Bush to get reelected.

8. Brian J. says:

I don’t know what to make of this, because the professor is an academic, so he probably wants the Democrats to win, and he’s from Green Bay, so he probably wants the Green Bay Packers to win when they play the Redskins on October 31 (Schedule).

Smells like a conspiracy of disinformation to keep the Republicans home when the Packers do, in fact, win.

9. As a HUGE Packers fan, I confounded. What to do? I am thinking of trading teams with Oliver Willis for one week.

## Fun with Stats

Apropos a discussion on the effects of the Bush tax cut at FactCheck.org and Begging to Differ, Chris Lawrence gives us a review of two often-confused measures of central tendency, the mean and the median.* It’s rather common practice for sides in a debate to pick the one that best fits their argument. The press typically doesn’t help matters by using the word “average,” usually understood to be a synonym for arithmetic mean, carelessly.

I agree with Chris that the median is almost always a more useful indicator, since the mean can be skewed by huge figures even in a large sample. This is especially true for income and similar economic variables, which are bounded at the bottom but not at the top. Even with something as large as the U.S. population, a handful of billionaires can skew the stats–whereas even the poorest of the poor have only a miniscule effect.
(more…)

FILED UNDER: Politics 101
James Joyner is Professor and Department Head of Security Studies at Marine Corps University's Command and Staff College and a nonresident senior fellow at the Scowcroft Center for Strategy and Security at the Atlantic Council. He's a former Army officer and Desert Storm vet. Views expressed here are his own. Follow James on Twitter @DrJJoyner.

1. I mentioned the mode in the second (and horribly long) footnote. I didn’t talk about geometric or harmonic means, though, which also have their purposes (generally as more robust alternatives to the arithmetic mean when you have outliers). And modes have their uses, but mostly as descriptive things (“I have a bimodal distribution” and such); also, the classification accuracy of some limited-dependent variable models can be measured versus preponderance of the mode.

I think the reason why the arithmetic mean is popular is (a) it’s the one people learn first, and (b) it’s probably the easiest to figure out, particularly if you don’t have a computer. And they work great when things are close to normally distributed or you can do the old asymptotic-handwave routine.

Although Statistics has always interested me, I’m pretty much stuck in the “average” crowd. And you definitely got me with the “asymptotic-handwave routine” phrase, Chris. I just have to nod my head and say, “He must know what he’s talking about, ’cause I sure as hell don’t!”

Oh, and to the point at hand: I agree with FactCheck.org on this point. Which figure is more meaningful, the average tax cut or the tax cut for the average taxpayer? It seems to me that the median is more realistic, but then again, it depends on what point you’re trying to get across. It isn’t “wrong” to use either figure, although it would probably be wrong to try to suppress one of them.

Where I differ with FactCheck.org is their insistence on including people who don’t pay income taxes in the calculations. While that can provide useful information, I’m personally not interested in giving an income tax cut to people who don’t pay income taxes, so the more meaningful figures, to me, are those that include only people who actually got a tax cut.

The payroll taxes are a separate issue, and being a bit anal-retentive myself, I think it would be better if the President would always say “income tax cuts” instead of “tax cuts,” and similar formulations. With his more stereotypical Texas cavalier approach to speech, though, I think that’s unlikely to happen. (I’m a Texan, by the way, so I’m speaking about my own group, thankyouverymuch.)

## FUN WITH STATS

Matthew Yglesias makes an interesting argument about a statistic he found in the LA Times:

In every election since 1960, the party in the White House lost when the unemployment rate deteriorated during the first half of the year. If the rate improved, the party in the White House won.

Says Matt,

But how sure are we that it isn’t a coincidence. 1960 was a long time ago, but we’re talking about a sample of just 11 elections. You wouldn’t do a public opinion survey of eleven randomly selected people and believe that there was any statistical validity to your results.

Well, when your entire universe is eleven, a sample of eleven is quite good indeed!

Matt’s point is a good one, though. The problem isn’t just that the number of elections is too small and the number of variables that could reasonably be expected to impact the outcomes too large to draw many hard conclusions. Analysis is further complicated by rather significant changes in the nature of the contests.

For example, Matt points out that 1960 seems an arbitrary starting point for this “study.” But, as we know, 1960 was the first real “television election.” We haven’t elected a short, bald, or bearded president since then. The major ideological battles, campaign finance rules, and the nominating process have all undergone numerous and significant changes over the years.

Also, while it certainly doesn’t prove anything, I would also note that the political parties don’t seem to believe in economic determinancy. If they did, one would think they would always rally behind their most ideologically pure candidate rather than trying to find an amiable centrist. If the election is simply a referendum on the economy, then ideology is irrelevant and there’s no point in selecting a compromise candidate.

FILED UNDER: US Politics