Rosh, Rush

Update:Uh... Jesse Taylor has already disemboweled Mary Rosh. Read him instead, or first.

Mary Rosh and Rush Limbaugh share something more than screwing themselves through the use of e-mail. They are both mortified that P.C. America no longer allows for uninhibited discussion about race in sports.

And even though Rush never defended Mary when s/he was under the gun, Mary has leapt to Rush's defense. Sure, it's a stupid defense, but what else would you expect?

Here's Rosh on Rush, in National Review:

To measure positive news coverage, I quickly put ten research assistants to work on a Nexis search, which is a computerized search of newspaper stories across the country. They looked at the coverage received by the 36 quarterbacks who played during the first four weeks of this season. (The articles were from the day of their first game to the day after their last game during the period.) To try to make the categorization of news stories objective, 23 phrases were picked to identify positive descriptions of a quarterback and 23 phrases for negative ones. Positive phrases included words such as "brilliant," "agile," "good," "great," "tough," "accurate," "leader," "intelligent," or "strong arm." Negative phrases included "overrated," "erratic," "struggling," "bad," "weak arm," or "mistakes." Obviously the media involves more than newspapers, but this is measurable and it is not clear why newspapers would be so different from the rest of the media.

Right away, the stupidity is stunning. Rosh has 10 researchers s/he can quickly mobilize to run Nexis searches on adjectival descriptions of quarterbacks? What conservative idiots financed this boondoggle?

Then there's the insipid premise of the survey -- that selecting 23 adjectives of praise and 23 adjectives of condemnation at random will yield an objective or accurate result of press attitudes. Certainly there are hundreds of other adjectives of praise and criticism, even in the vocabulary of sportswriters, and those articles go uncounted with no way of knowing whether more pro articles or con articles were omitted.

We then identified news stories where these phrases appeared within 50 words of a quarterback's name. Each story was read to check that the phrases were indeed used to describe the "quarterback" and to make sure that the word "not" did not appear before the different phrases. Depending on whether positive or negative words were used to describe the quarterback, stories were classified as positive, negative, or falling into both categories.

If you're going to read all the stories anyway, and only take a four week sample, why not read all stories addressing quarterbacks and see if they include a positive or negative characterization?

The evidence suggests that Rush is right, though the simplest measures indicate that the difference is not huge. Looking at just the averages, without trying to account for anything else, reveals a ten-percent difference in coverage (with 67 percent of stories on blacks being positive, 61 percent for whites).

We also collected data by week for each of the first four weeks of the season on a host of other factors that help explain the rate at which a player is praised: the quarterback's rating for each game; whether his team won; the points scored for and against the team; ESPN's weekly rank for the quarterback's team and the opponent; and whether it was a Monday night game. In addition, I accounted for average differences in media coverage both in the quarterback's city and the opponent's city as well as differences across weeks of the season.

And what the hell did Rosh do with all of these variables? Did Rosh decide whether the negative or positive description was warranted? If so, what was the criteria for her determination? The rate at which a QB is praised is explained by whether he played in a Monday night game? This is so completely unscientific it's laughable.

Accounting for these other factors shows a much stronger pattern. Black quarterbacks' news coverage is 27 percentage points more positive than whites. And that difference was quite statistically significant � the chance of this result simply being random is the same odds as flipping a coin five times and getting heads each time.

The quarterback ranking, scoring, winning, and higher-ranked teams playing against each other all increase the percentage of positive stories.

Huh? What the hell does that mean?

For example, each additional point scored by the quarterback's team raises the share of positive news coverage by about one percentage point. Being in the only game played on a particular day lowers the how positive [sic] the coverage was by about 12 percentage points, as more newspapers outside the home area cover the game the next day.

So how did Rosh use these factors in determining whether the black QBs were overpraised? Beats me. Rosh certainly doesn't explain it.

Without giving this stupid article more time that it deserves, it is sufficient to say that Rosh's statistical gibberish doesn't show that "the media" overrates any black QB, let alone that it overrates any black QB because of his race. And that was Limbaugh's claim. In the end, Rosh provides Rush zero vindication.

I sure hope Rush gets a better defense from Roy Black.