End-of-term Catch Up: Teach Your Kids to Throw Left-Handed

The end of the academic term is M-U-R-D-E-R on the baseball commentary.

Measuring the Corps of Relievers, after 14 April 2010. (Yes, you remember - Philadelphia 14 — 7 Washington. And Bergmann got the Black Spot.)

Matt Capps      .352
Tyler Clippard  .188
Brian Bruney   -.056
Jesse English  -.074
Tyler Walker   -.123
Miguel Batista -.259
Jason Bergmann -.485
Sean Burnett   -.509

Those numbers are the cumulative amounts of win probability added or deducted during all the individual relievers' stints between Opening Day and 14 April. Really, Burnett, loser of half a game all by himself, was the man doing the least. (Plus he was more evenly bad, while Bergmann loaded all his bad into one outing.)

But Burnett's left-handed.

So boys and girls, remember — baseball is more forgiving of left-handed pitchers.

What a difference a year makes?

From 1-10 to 6-5. .500 baseball. Huzzah!

Not to rain on everyone's parade, but this year's start puts last year's in perspective:

          Runs Scored     Runs Allowed
2009         54              75
2010         53              66

The difference in runs allowed would actually only add up to about half a win.

QMAX is a system from the old Big Bad Baseball Annuals of adding some 'granularity' to the Quality Start, showing that some Quality Starts are better than others. 2009:
2010:

Click on the images to make them bigger. The blue area marks the 'Success Square' outings, where the combination of hits allowed and walks surrendered is good enough to mean you should win half the time. The red zone shows the 'Hit Hard' starts, games one is unlikely to win. There's not a whole lot of difference between this year and last: one fewer hit hard and an elite quality start, and that was Livo!'s of yesterday. That one also probably skewed our runs allowed statistic.

The 2009 edition of the Nationals underperformed badly. The 2010 one has overperformed slightly. You could put it down to Mr Rizzo's choice of 'winners' versus Bowden's 'lollygaggers'. You could put it down to a firmer hand on the tiller than last season, when Mr Acta seemed to be on borrowed time from the autumn, and MLB wouldn't let Mr Rizzo shed the 'interim' tag. You could put it down to luck. I know which one I think. How about you?

Dodgy Data?

Colin Wyers' has written an interesting article at BaseballProspectus.com about differences in baseball batted-ball data depending on who is assembling it. Since I never regarded batted-ball data as being anything better than an approximation of longitude while at sea without a chronometer, I'm not undergoing quite the crisis of faith that Wyers is. Nonetheless, it is worth bearing in mind when reading about sabermetrics based on batted-ball data.

3,000-and-some words on Garrett Mock

Garret Mock was sent down to AAA Syracuse today to make room for Livan Hernandez. At first, this move sent me into a momentary despondency. I thought he would get five or six starts, and the final decision of whether he has a future on the Nationals' big-league staff could be made. One start is not enough. Mock was being handled very poorly, in spite of his pitching shortcomings. This is the Big Moan I have about the Nationals, that they seem to handle personnel decisions very poorly.

The more I thought about it, though, the more it made sense in terms of roster strategy. The Phillies just showed us last week we need plenty of relievers. Furthermore, there are good reasons not to give up hope on Garrett Mock. However, pictures speak a thousand words so I'm going to show you a picture:

Then I'm going to show you another picture:

And finally a third picture:

(Images courtesy of the wonderful Brooks Baseball.)

Everybody says Mock has got good stuff. These three pictures highlight what that means in practical terms.

The top one shows the location of Livo!'s pitches in his start yesterday. The middle Stammen's Thursday start. The bottom is Mock's last start for the Nationals.

First I draw your attention to the blue rectangles. These represent the case for Mock. Stammen can't beat bats in that zone. Livo! avoids it altogether. Mock can pitch in that zone, and beat bats. If you study these three diagrams you'll see that batters do best at getting hits in that zone. Mock can challenge hitters; Stammen needs the batters to make weak contact so his fielders can get him the outs. It's much safer to beat hitters, if you can.

Next, I point you to the red circles. What on Earth are all those pitches from Mock doing there? They have no apparent relationship to the strike zone. Note how both Livo! and Stammen managed to keep almost all their pitches closer to the strike zone.

Mock has his objective very clear down in AAA. Stay out of the red circle.

It's the Lineup?!?!

After handing out a Hero award, I didn't have to wait long at all to hand out a Goat one. Jesse English claimed it, in spite of Tyler Walker making a late run at it.

Why not Batista, who got the loss?

The reality is that the real goat yesterday was the lineup. (You could also argue that Mike Pelfrey, whom I thought pitched very well yesterday, won this game.) Batista gave up one run. The lineup had two chances to get that back. In the top of the fifth, the heart of the lineup went meekly 1-2-3. In the top of the sixth, we had Willingham on third with one out, and neither Pudge nor Desmond could score him.

In the end it made no difference, because Bergmann and English gave up two hits that increased the Mets' lead to two runs, then Walker came out with a flamethrower and an entire P-O-L dump to put the game out of sight.

I'm looking at the Nationals' lineup with some concern now. It's not really working, and it doesn't help that Kennedy is in a funk. This was supposed to be our strongest piece, and it's just not helping.

First Bullpen Hero Award for 2010

Last year my frustration with the bullpen drove me to create the Bullpen Goat of the Day awards, which I later in less churlish mood supplemented with Bullpen Heroes of the Day. I fully anticipated that the first award of 2010 would be a Goat, but Matt Capps managed to snatch the Hero's palm away from Brian Bruney yesterday.

These awards are based on the concept of 'P', the work of sabermetrician Doug Drinen which first appeared in the 1990s-era Big Bad Baseball Annual (previously known as the Baseball Sabermetric and the lineal line of descent from Bill James' original Baseball Abstract series). 'P' calculates the win probability at the moment when a reliever enters the game, and what it will be if he retires the side from that point. I've added the actual result of the pitcher's work, and calculated the difference between the ideal and the actual. Here are the scores from yesterday's game:

Capps         .175
Bruney        .127
Clippard     -.004
Burnett      -.352

Another statistic I have lifted from the Big Bad Baseball Annual stable is QMAX, which is a more sophisticated way of analysing starts than the Quality Start/Not Quality Start binary opposition. The interesting thing is that Craig Stammen pitched the worst kind of start (in the 'Hit Hard' category), and yet the Nationals' still won. That means the lineup and the bullpen were the difference in this game. And with a 1-2 record after three 'Hit Hard' starts in a row, against the tough Phillies' lineup, it shows that the despair that many may have felt going into yesterday's game might be a trifle misplaced.

Astonishing! Yes, It Is!

When Cristian Guzmán took over from Mike Morse in rightfield last night, I was astonished. Yes I was. Weren't we told Guzmán's arm wasn't up to playing at shortstop? It is actions like this that give sober men pause. Now it seems that according to an Adam Kilgore 'tweet', Desmond-of-the-two-errors is going to make way for Guzmán from time to time.

You could look at Fangraphs win probability chart and see that Josh Willingham was our batting star of the game. Or you could look at my own spreadsheet, with my own version of leveraged win value (which cumulatively appears in the sidebar), and see that Josh Willingham was the batting star of the game. This gives me confidence in my own system.

Finally, I'll just comment that despite Federal Baseball's headline, Marquis' outing was actually worse, by my measure, than Lannan's.