Michael Steele, former chairman of the Republican National Committee, appears to be on a self-promotional tour focused on insulting current RNC chair Reince Priebus. Today on MSNBC, he argued that "I won, and he didn't." Which, okay, sure, except Michael Steele had nothing to do with the election he won.

Over the weekend, Priebus offered a vision for how he intended to rebuild the party after a rout in the 2012 elections. His focus was on message and demographics, which led others to suggest that he was misreading the problem. Phyllis Schlafly criticized the candidates, which led Steele to tweet:

Earlier today, Steele continued the attack on MSNBC.

Casting occasional sideways glances at the camera, Steele said:

I won, and he didn't. We laid down a ground game, national fifty state strategy. We didn't have to go through the hoopla of press conferences. We just did the heavy work of rebuilding the party, coming off of massive losses of 2006 and 2008.

To which Priebus could offer a two-word response. Or one word, if you spell "bullshit" without the space.

Michael Steele's reasons are not the reasons why Michael Steele's Republican party won in 2010. He "won" because he had a much more favorable electorate — the sort of demographic shift which even a robust voter turnout operation, which Steele didn't have, couldn't have created.

Looking at the exit polls for 2010 and 2012, the shift is immediately obvious.

The blue lines are turnout percentages in 2010; the red, 2012. If a blue line is taller than the red one, it means more of that demographic turned out in 2010. Or, to put it another way, here's the change in turnout percentage between 2010 and 2012.

The differences: Men, Republicans, whites, older people, conservatives. Think that helps explain the margin of victory?

You might rightly ask if those differences were a function of Steele's wicked awesome turnout tool. Those familiar with campaign "ground games" would recognize that you're not going to do 22 percentage points better with voters over 65 no matter how excellent your field work might be. Doing five points better would be a miracle.

But the argument is actually simpler than that. Here's the ideological breakdown applied to the total voter pool (2010 data, 2012) in each year.

The GOP lost in 2012 because far more liberals and moderates (who voted for Obama) came to the polls. But so did more conservatives! Some 6.2 million more conservatives voted last year than two years prior. In other words: Priebus' Republican party did a better job getting people to the polls.

No one should be surprised that Steele's 2010 election went better than Priebus's did in 2012. (Priebus took the chairmanship in January 2011.) As Michael McDonald indicates in the white paper, "Voter Turnout in the 2010 Midterm Election," midterm elections have seen significantly lower turnout for almost 180 years. And lower turnout these days favors older, whiter voters.

It's understandable that Steele would want to repair his legacy. Before now, he was best known for leaving the party $23 million in debt at the time he left his position. Between that and today's faulty argument, perhaps we've put our finger on the root of the problem: Michael Steele isn't very good at math.