Saturday, June 11, 2005

The Curse Of Joe Horgan

As the Nats have continued to pile on the wins (what are we at? 43 in a row?), the little devil has been sitting on my shoulder whispering doom and gloom scenarios.

The major reason for that has been our runs scored and runs allowed

Bill James, the Grand Poobah of baseball statistics, came up with a quick and dirty forumula which quanitifes the relationship between runs and a team's record: his Pythagorean Theorem.

All it says, in effect, is that a team that outscores its opponent will have a better record than one that's outscored. Makes sense, right?

It has several implications, though. And it builds on one of the tennents of stathead type thinking; one-run games are mostly a function of luck.

Heresy, you say! (And you're probably right to some degree), but if you look at the one-run record of teams, the teams with the best records don't necessarily have the best one-run records. And the KC Royals of the world don't always have the worst record.

Which, if you stop to think about it, probably makes sense.

If the Kansas City Royals are playing the St. Louis Cardinals and the Royals win, what do you think the final score will be? 2-1? 4-3? Maybe they'll go crazy and win 5-3. But what do you think a Cardinals victory would look like? 10-2? 8-3?

It's much easier to envision the Cardinals blowing out the Royals than the opposite. If the Royals are going to win, they're going to probably eek out a victory.

So, while the Royals may eek out a 1-run game against the Cardinals, chances are the Cards will beat the crown off their head in the other two games they play -- a result that's probably closer to the true level of ability for these two teams.

So, I sit here this morning, noticing that the Nationals Runs scored finally exceeds their runs allowed (255-252).

That's a result that, according to James' formula, would yield a roughly .500 record. Instead, we're well above .500 (nine games now!)

And, I look at the stretch of games over the last two weeks, and I see a lot of one-run games -- games where, if we didn't get that ONE big late-inning hit, we'd have lost.

And it makes me nervous.

I hate pythagoras!

But, then, I start to think about it a little bit.

First, is Nats Blog's look at Pythagoras from earlier this week, using the situational dependent WV. Although the point of the Theorem is that WHEN runs score (over the course of a loooong season) isn't as important as how many are scored, theirs is an interesting look at how our 'clutch hits' indicate that we're on the right track.

Second is the Joe Horgan factor.

You might recall Joe Horgan, the second left-hander in our bullpen at the open of the season. He's the guy that Frank wanted to have on the staff so badly, he forced Bodes to send Gary Majewski down to the minors. Yeah, that Gary Majewski. Good call, Frank!

Our friend Horgan pitched like he was an agent provocateur. I know I certainly felt violent watching him pitch. And Frank did his best to kill him (and the team) by leaving him out there to be brutalized by opposing batters.

When the smoke cleared, Horgan had left a trail of 14 extra runs.

His accomplice in provoking, Antonio Osuna yielded 11.

Both those pitchers were at their most ineffective in losing efforts. They turned raging out of control bonfires into Chicago circa 1871.

Well, what if the root of the Nationals' problems was Joe Horgan? What if Horgan was the equivilent of the Brady Bunch's Hawaiian Tiki?

I separated the season out into two parts: the dark ages WITH Horgan, and the reformation, post-Horgan.

With Horgan, this team was outscored 69-90.

Without Horgan, we've outscored the other team 186/162.

Plugging those post-Horgan numbers into James' forumla yields a nifty .569 winning percentage, which produces 35 wins. Our actual total? 35!

For the last two months, We're outscoring our opponents pretty handily, and we rarely are blown out. (Of course we rarely do the blowing either) But, Sans Horgan and Osuna, things are looking up.

I'm relieved now. The Curse of Joe Horgan has apparently been lifted, and we're free to win with impunity.

5 Comments:

  • Does that make Frank ... Greg Brady?

    By Anonymous Anonymous, at 6/11/2005 10:10 AM  

  • Given how Horgan reported some slight arm problems, he's probably more like Sam the Butcher! :P

    By Blogger Chris Needham, at 6/11/2005 10:17 AM  

  • Yeah, I know. I was half being facetious, half being realistic.

    My larger point, which is probably lost in the 2,500 words is that this team HAS been playing up to its record over the last two months.

    We've already beaten Pythag from early in the season... We're playing with a clean slate.

    By Blogger Chris Needham, at 6/11/2005 4:10 PM  

  • Good post, and thanks for the cite. As I looked at Win Value more closely, though, I'm now convinced the gap between Pythag and Reality is not clutch hitting, but clutch pitching (i.e. bullpen, Livan, etc.) So you're right to point out the Horgan factor, and it also bodes well for our future, because I think our offense has underperformed this year, on the whole.

    By Blogger DM, at 6/12/2005 12:42 AM  

  • I was on vacation over the weekend and just now got to see this. I think you can't relate the winning percentage based on the non-Horgan era to the entire season like you are doing.

    With Joe: 18 games, 73-91 (I'm pretty sure about this), .391 winning %-age, 7 wins predicted 9 actual

    Without Joe (up until your post): 43 games, 182-161, .561 winning %age, 24 wins predicted, 26 actual.

    So 31 predicted wins total, 35 actual. No matter how you slice it we're still overperforming.

    By Blogger Harper, at 6/14/2005 3:22 PM  

Post a Comment

<< Home