Numbers as Narrative: Bell, Hernandez, Pujols

Scott Gray’s The Mind of Bill James is an excellent book — one whose entire contents I’m tempted to reproduce here except for (a) it’d break every copyright law in existence and (b) all that typing would hella exacerbate my ulnar claw.

Anyway, believe me when I say that Gray does an excellent job of demonstrating the wide-ranging import of Bill James’s work. Believe me, also, when I say that Gray is smart to occasionally quote at length from the Abstracts and other of James’s books.

Which, that’s exactly what he does at the beginning of chapter three of the aforementioned book. Regard, the Master at work:

When the numbers melt into the language, they acquire the power to do all of the things which language can do, to become fiction and drama and poetry. Am I imagining things? Do not the numbers of Ted Williams detail a story of fierce talent and, by the char of their ugly gaps, the ravages of exquisite frustration that ever accompany imperfect times? Do not the numbers of Roberto Clemente spell out a novella of irritable determination straining toward higher and higher peaks until snapped suddenly by an arbitrary, but now inevitable, machina? Do not the stressed and unstressed syllables of Willie Davis‘ prime suggest an iambic indifference Is there not a cavalcata in Pete Rose’s charges? Is there no union of thrill and agony in Roger Maris‘ numbers? How else can one explain the phenomenon of baseball cards, which is that a chart of numbers that would put an actuary to sleep can be made to dance if you put it on one side of a card and Bombo Rivera’s picture on the other.

First off, I’d like to mention how, despite being alone in my living room when I first read that, I still stood up and began a Hollywood movie-style slow clap.

Second off, here’s the exciting idea that James gives us: that numbers are able to take on a narrative quality.

My reaction when I read that was something like: hey, I wanna play along. Lucky for me, I have a platform to do that.

So what I did was to isolate three players whose numbers tell an interesting story. Here they are.

Heath Bell, RHRP, San Diego
The story that Bell’s numbers tell us is of a pitcher whose talent was, in the parlance of our 43rd president, misunderestimated. The guilty club? Shockingly, the Mets. Yes, the same team whose bullpen effectively handed division titles to Philadelphia in both the 2007 and 2008 seasons didn’t have space for a pitcher who was posting good fielding-independent numbers. Regard Bell’s first three MLB seasons:

Season	Team	IP	gmLI	ERA	tRA	tRA*	FIP	xFIP	BABIP
2004	Mets	24.1	0.93	3.33	3.81	4.08	4.24	3.02	0.290
2005	Mets	46.2	0.75	5.59	4.01	3.98	2.91	3.39	0.374
2006	Mets	37.0	0.59	5.11	5.69	4.60	4.25	3.21	0.394

That gmLI thing, in case you don’t know, is the average Leverage Index of all the times a pitcher enters a game over the course of a season. One (1) is average. Below one indicates a low leverage situation. What those declining gmLIs show us is a team (New York) becoming increasingly hesitant to hand over important situations to one of its relievers (Bell). And if you look only at Bell’s ERAs, you’ll understand why. Were Bell’s 2005 and 2006 ERAs indicative of his skill, then New York’s reluctance would make sense. But cast your eyes rightward to the end of those rows, and you’ll see the more likely culprit for Bell’s seeming mediocrity: way inflated BABIPs. The most likely reason for Bell’s poor showing was a combo deal of bad luck and shoddy D. His tRA of 5.69 in 2006 is a red flag, but the regressed version (tRA*) from StatCorner and all the other sorts of metrics all say Bell wasn’t far off from his established levels.

In Novemeber of 2006, San Diego GM Kevin Towers sent Ben Johnson to New York for Bell and Royce Ring*. Johnson disappeared after 2007**. As for Bell? Ta Da!:

Season	Team	IP	gmLI	ERA	tRA	tRA*	FIP	xFIP	BABIP
2007	Padres	93.2	1.41	2.02	2.57	3.01	2.50	2.84	0.260
2008	Padres	78.0	1.57	3.58	3.62	4.18	3.34	3.87	0.291
2009	Padres	64.0	1.83	2.81	2.58	3.42	2.53	3.17	0.294

As you can see, he proceeded to be awesome. Awesomer than before, in fact. Whether that’s luck or actual improvement or because he had an established role, we don’t know. The point is, the talent was there. And even if said talent resided/resides in a sort of overweight-looking manbody, if said manbody strikes batters out and gets groundballs, who the eff cares?

*Setting off a nightmarish chain reaction of “Ring that Bell” puns across all America’s sports pages.
**Seriously. No dice from any of Baseball Reference, Baseball Cube, or Wikipedia. Is he okay?

David Hernandez, RHSP, Baltimore
Let’s play a game. The game is called One of These Things Is not Like the Other. The first thing you do is look at some numbers. Second, you say which one doesn’t belong. It starts right now.

Season	Team		IP	SO	K/9
2006	Orioles (A)	145.1	154	9.54
2007	Orioles (A+)	145.1	168	10.40
2008	Orioles (AA)	141	166	10.60
2009	Orioles (AAA)	57.1	79	12.40
2009	Orioles		91	59	5.84

Okay, so it’s not a super fun game, but at least it’s easy. What you see there are David Hernandez’s raw stats from 2006 – 2009. What you’ll notice is how is K/9 rates are super-good… in the minors. The Show hasn’t been very kind to Hernandez, right up to his most recent start versus Boston when he struck out only one in six innings.

Of course, the Average Reader is smart enough to say, “Well, sure, but those are just his raw minor league stats. Obviously his major league K-rates will be lower. The question is, How much lower?” Luckily for all of us, Jeff Sackmann’s on top of that junk. According to Minor League Splits, those same seasons (minus the 2009 MLB one) look alot like this:

Season	Team		MLE IP	MLE K	MLE K/9
2006	Orioles (A)	108.2	69	5.74
2007	Orioles (A+)	149.2	119	7.18
2008	Orioles (AA)	135.1	128	8.53
2009	Orioles (AAA)	57.2	66	10.38

It’s interesting to note how dramatically Hernandez improves across these four seasons. Speaking anecdotally, I’m not sure I’ve seen many pitchers exhibit such stark linear improvement. Usually you get some regression here, some break-out there. But those K-rates increase almost exactly 1.5 K/9 every year. It’s strange.

But I digress. Back to the point: What’s the story here?

Well, there are two, kinda. The first concerns Hernandez himself. It’s a story we’ve heard before, as Dave Cameron pointed out a couple months ago: a pitcher dominates the minors with deception and a good breaking pitch, gets promoted to the majors, and gets shelled. Yusmeiro Petit‘s story is similar. So is Brandon McCarthy‘s and Garrett Olson’s.

The other story is of the mystery genre. It’s not quite Poirot-level excellent, but pretty good. It concerns a whole team of scientists (read: sabermetricians) who are unable to find a key missing ingredient to solve a case. The “case” requires that they project a pitcher’s potential major league success based on his minor league numbers. What they (i.e. the scientists) have noticed is that some pitchers simply don’t pitch at the major leagues like you’d expect them to.

Meanwhile, a group of gritty, skin-of-their-teeth detectives (read: scouts) say it’s because those certain pitchers don’t have major league stuff. The scientists say…

Okay, I’m gonna stop that, for all of our sakes. Here’s the point: Hernandez is young and his performance could improve, but the fact remains that he’s fairing at the major league level almost exactly how his scouting reports and pedigree suggested he would. Fin.

Albert Pujols, Crusher, St. Louis
El Hombre’s numbers tell the story of a young Dominican who immigrated with his family to the US, attended Maple Woods Community College in 1999, and either during his stay there or at least some time before 2001, BECAME A BASEBALLING CYBORG.

There’s no other possible explanation for Pujols’s twin traits of durability and production:

YR	PA	wOBA	Rank (Among players with > 400 PA)
2001	676	0.421	12
2002	675	0.402	18
2003	685	0.462	2
2004	692	0.439	3
2005	700	0.436	3
2006	634	0.448	2
2007	679	0.414	13
2008	641	0.458	1
2009	637	0.458	1

It’s hard to divine exactly what sort of transformation Pujols might’ve undergone — more like a RoboCop/Megacorp dystopian law enforcement sort of scenario, or more like a Wolverine/Weapon X clandestine government op-type of situation.

The fact that ESPN has attempted to make light of Pujols’s “secret” via one of their light-hearted This Is SportsCenter commercials should only serve to (a) strengthen our conviction that Pujols is indeed half-man, half-machine, and (b) suggest that ESPN — and, by extension, Disney Corp. — is somehow behind this operation.

My guess? That Pujols’s body was originally constructed to replace the current Teddy Roosevelt in the Hall of Presidents, except then the Disney execs realized that Teddy Roosevelt wasn’t shaped like the Incredible Hulk.