Wednesday, March 31, 2010

F@$% This!

A few years ago, I wasted about a half-hour out of my day in creating a junk stat called Fielding- Unassisted Precisionlessness, or “F-UPs” as it would have been commonly known if it for some strange reason had ever become commonly known. Actually, to say that I created this junk stat might stretch the truth somewhat; all I did was cobble together a dumb “metric” with the aid of some established statistical categories. I did, however, create the name of the junk stat, in so doing paddling fearlessly against the course of the “IPs” revolution that dominated stats-nerd nomenclature during the Aughts (DIPs, FIPs, ZiPS, and so forth).

To put it simply, F-UPs is a pitching stat that expresses a team’s combined walks, wild pitches, and hit-by-pitches per nine innings pitched. It’s designed to be a quick-and-dirty way of determining whether a team’s pitchers have, generally speaking, any idea what the hell they’re doing out there. Call it the Moron’s ERA.

When last I visited the subject, early in the 2007 season, the Nats were pretty notorious eff-ups, so to speak. Nats pitchers in 2006 had posted over an F-UP per nine innings greater than the NL average, and 2007 wasn’t off to such a hot start. As it turned out, the Nats’ F-UPs exceeded the NL average in both 2007 and 2008. And 2009 certainly wasn’t great in that regard either.

In 2009, the average NL team pitched 1444 innings, issued 564 walks, hit 53 batters, and tossed 55 wild pitches. That comes out to 4.19 F-UPs per nine innings.

(By the way, the average NL ERA last year was . . . 4.19. Ha!)

By comparison, in 2009, Nats pitchers threw 1424.1 innings, with 629 walks, 48 hit batsmen, and 74 wild pitches. Thus, the Nats recorded 4.75 F-UPs per nine innings, well above the league average yet again.

Not to belabor the point, but it’s rather evident that the Nats don’t have a great recent track record according to my junk stat. This might tend to reinforce the impression, corroborated by the won/lost record, that the Nats haven’t been especially good. While it’s certainly not impossible for a team to thrive despite a bad F-UP tally (the Dodgers, who exceeded the league average in all three components, also had the best record in the NL), you’d think that high aggregate totals of BB, HBP, and WP would generally reflect a team lacking the essentials.

Of course, F-UPs has its limitations, as you’d expect from a junk stat. First, it’s not exactly a “fielding-unassisted” measure; the difference between a wild pitch (on the pitcher) and a passed ball (on the catcher) isn’t always clear. Second, the degree to which managers feel comfortable issuing intentional walks is certainly not uniform. And third, given that there are ten times as many walks as hit-by-pitches or wild pitches, it’s largely an overly-elaborate way of expressing a team’s walk rate; accordingly, if a team (like the Nats) tends to exceed the average walk rate, it will tend to have an above-average (below-par) F-UPs figure. Just so we’re clear, therefore, I’m not pitching this idea to Baseball Prospectus.

But the fact is that the Nats have consistently sucked by this measure, as evidenced by their outrageous walk total last year. As noted above, walks are the driver in the F-UP equation. Oh, those walks. Last year, the Nats issued 629 walks, a silly total, the most in the NL by far. John Lannan led the team with 68 walks, but he certainly wasn’t the problem. Rather, it was a drip-drip-drip of errant tossers who contributed to the excessive total – Wildassery-by-Committee, if you will.

Nobody horrific stayed around too long, but there were a lot of horrific performances in rather short but stinky bursts, so combinations of them serve to demonstrate how much these guys skunked up the works. Add Daniel Cabrera (35 BB in 40 IP) to Garrett Mock (44 BB in 91.1 IP), and you get 79 walks in 131.1 innings from what was essentially the fifth slot (I guess?) in the starting rotation. Perhaps that’s not fair to Mock, but whatever. Add Mike Hinckley to Ron Villone (who replaced Hinckley), and you get 41 walks in 58.1 innings of lefty relief. Add Julian Tavarez to Kip Wells, and you get 45 walks in 61.1 innings of righty relief. That’s just scratching the surface.

The bullpen, in particular, had some guys with really bad walk rates. My impression is that Rizzo likes pitch-to-contact guys in his rotation, but will tolerate relievers who issue walks. A lot of last year’s junk is gone, but he brought in Brian Bruney and brought back Mike MacDougal. Regardless, it’s a given that the Nats need to cut down on the free passes in 2010, and you don’t need a junk stat to reach that conclusion.

But the other components do add up. According to this site, for instance, the linear weight for each wild pitch (or passed ball) is 0.28 runs. Let’s run with that figure. The Nats were bad with wild pitches last year; their total of 74 was third-highest in the NL. Based on a league average of 55 wild pitches, the Nats were 5.32 runs below average in terms of wild pitches – about half the total of a mythical win. In terms of linear weights, the difference between the most wild pitches in the NL (Arizona, with 79) and the least (St. Louis, 28) was about a win-and-a-half.

* * * *

Okay, so we’re got (1) a high walk rate, (2) exacerbated by lots of wild pitches. Some thoughts:

- If you’re really bored, you can adjust F-UPs to account for intentional walks. The Nats issued 59 IBBs last season, fifth highest in the league, ten above the league average. Riggleman tended to be a bit more liberal with the tactic, with 33 in 75 games (as opposed to 26 in 87 games from Acta).

- This discrepancy might be explained to some extent by circumstance, but there’s some data pointing to differences in managerial philosophy. The Nats issued a below-average amount of IBBs in both of Acta’s full seasons. Riggleman’s teams were NL leaders in IBB three years running (1993-95). Although he had stepped off the pedal somewhat by the late 1990s, Riggleman has called for an above-average number every full season he’s managed. Not that we should do this, but the 33 IBBs under Riggleman’s watch, projected to a full season, would have led the NL last year.

- As it were, for the entire season the Nats issued 10 more IBBs than league average, and that figure accounted for about 15% of the difference between the Nats’ total walks and the amount of walks issued by the average NL team.

- You can also adjust F-UPs to account for Daniel Cabrera. In addition to the 35 walks in 40 innings, Cabrera uncorked 10 wild pitches. For the record, Cabrera managed to F-UP 10.58 times per nine innings.

- Nevertheless, the Nats had a lot of wild pitches even adjusting for Cabrera. They also threw a lot of wild pitches, relative to the NL, in 2007 and 2008. It is entirely possible that these figures are attributable to any or all of the following: a) really crappy pitching (a given), b) crappy catching, and c) dumb luck.

- To the extent the wild pitches are attributable to crappy catching, well, the Nats have had different primary catchers the past three seasons so that’s pretty diverse crappy catching. Then again, the Nats have been crappy in general (or worse!) the past three seasons, so it seems plausible that they got crappy catching too. Brian Schneider was pretty well regarded, but wasn’t 2007 the season he was alleged by one reporter to be “complacent”? (Maybe that was 2006.)

- The Houston Astros, on the other hand, had a comparatively low total of wild pitches in 2009, and Pudge Rodriguez was their catcher much of the year. But they also had a greater than average number of passed balls last year, which suggests that’s a matter for the official scorer. However, the league-wide number of passed balls is so low that the total number of wild pitches and passed balls by the Astros was still pretty low. Thus, to the extent a catcher’s defense effects the number of wild pitches thrown, it might be an interesting experiment to watch whether the number of wild pitches the Nats throw is affected by Pudge’s apparent regular playing time (at least until Jesus Flores returns, if and when).


Post a Comment

<< Home