Don’t Label Me!
A very good Matt Garza article on FanGraphs Wednesday ended up leading to a lively Twitter debate that I think is worth fleshing out without character limits in place. I’ll try not to simply pimp the virtues of my perspective at the expense of the alternative.
The exchange began with a tweet from Mike Ferrin, Sirius/XM radio host and co-host of Baseball Prospectus’s Fringe Average podcast. Within the article, author Jeff Sullivan had called Garza a “solid number-two starter,” and Ferrin objected to the designation, although really, just in passing.
That objection caught my attention. I have seen quite a bit of Garza, going back to his days with the Minnesota Twins. I doubt I know anything about him that Mike doesn’t, but for me, Garza is a three-win pitcher when healthy. He throws in the mid-90s, and his slider is undeniably plus. He strikes out a lot of batters, although not an extreme number, and he does so even while pitching in the strike zone, so he doesn’t walk many. I did a quick count, scribbling down names on paper, and came up with only 17 pitchers I find to be decisively better than Garza, with almost that many right on his level, too.
I tweeted back to Mike that I thought Garza was certainly a top-40 guy, wondering whether he felt otherwise. His response:
Scouts use the 1-5 designation. There are less 1’s & 2’s combined than 3’s total. So he can be a top 40 & a number 3.
That was what I had been afraid of. I’ve written about my distaste for the arbitrary, vague and subjective delineation system scouts use to project—well, I’m not really sure, since it clearly isn’t about roles, and is not specific about value; profile, maybe? Anyway, I don’t find that structure to be one that lends itself well to specific player analysis or valuation, so I always seek out more information when someone uses the numbers to describe someone.
Since Ferrin basically agreed on where Garza ranks among his peers, I could reasonably have left it there. But I’m a smart-ass:
Heh. Yeah, THAT’s why the 1-5 system is dumb. That doesn’t make sense. And don’t Goldstein me how it doesn’t have to.(R.I.P.)
(Can’t stop now to explain the joke at the end. Listen to BP’s old podcast, Up & In, to get it.)
It’s more about the qualities a pitcher has. I’m sorry you think it’s dumb. It’s just not gonna change.
And we were off. I reluctantly concede on the inevitability of the system. For better or worse (and it’s a bit of both, truth be told), scout jargon will always rule discussions between intelligent and dedicated baseball fans. It’s the language of the game. It marks the speaker as one of us, and it creates familiar archetypes that let the conversants talk about individuals in broad contemporary and historical context.
I said as much, too. I am not going to spend my life tilting at semantic windmills. That system is what it is, and although Colin Wyers jumped in to say:
I come from the school of thought that there’s 30 of each [type of SP, 1 through 5].
I don’t have the energy to join him in trying to reform it. Instead, I want to work around it. I want the discussions we have about pitchers, who are very different from one another, and about pitching, which is dleightfully, mind-bogglingly complex, to run much deeper, and even when they don’t, to be much more descriptive than the numerical system allows.
I do have a couple more things to which to object, though, before moving on. One thing Ferrin replied as the conversation progressed:
It’s about communication. Trying to change the terminology, leads to confusion. Better to accept parameters & discuss within.
didn’t ring true at all for me. I can’t see that it’s better to fall into that terminology and let it drive a debate, because the result thereof will often be a divide over which category a guy falls into. It’s fine to let the parameters of the ratings lead a discussion, but one must remember that those numbers aren’t independently significant or meaningful.
It doesn’t make any difference in the player’s actual, absolute value whether he’s a 2 or a 3—that’s only a distinction made for its own sake. It also hems in the conversation. You lose the ability to debate small elements, small differences, because how does that impact whether he’s a 3 or a 4? It doesn’t. Which is part of the point.
Another point: Mike went on to talk about how the designations aren’t about slots or roles, but evaluations of talent. It has to do with the number of above-average pitches in the arsenal; the command of each pitch; durability; and pitchability.
The thing about all of that, of course, is that pitching is very much a matter of the whole, not the sum of parts (even if one of the parts being measured is how they synthesize into a whole). Moreover, pitching is just not this easily reduced to summations like these. Pitchers’ stuff and mechanics vary from start to start, month to month, year to year. Pitches come and go from repertoires. Interdependence abounds, from the catcher to the umpire to the defense. I just don’t see that using a scouting shorthand describes a pitcher’s value as well as taking the few extra sentences to break down how they do what they do.
Travis Wood Just Keeps Pitching
Another thing I tweeted this morning was that the consistency of Travis Wood’s 2013 seems like it should be impossible. I wanted to expand on that here, too.
Wood has made 18 starts this season. Seventeen have met the minimum standard of a quality start, meaning Wood went at least six innings and allowed three or fewer earned runs. (He did allow four total runs in one of those outings.)
His average Game Score (a Bill James construction wherein a pitcher starts with a score of 50, and is awarded and debited points based on runs, home runs, hits, outs, walks, strikeouts, you get it; 55 is average) is 60, but the astounding stat is this: The standard deviation around that just a shade over eight. Wood has had three very good starts, and just one bad one. Fourteen of 18 starts have been, basically, average. That’s not sexy, but the regularity of it is wildly valuable. After all, a league-average pitcher is a valuable thing, and if you don’t have the blow-up starts to which so many guys are prone, the team always has a chance to help you overcome a mis-timed mistake.
It’s not just the absolute value Wood delivers each time he takes the mound that’s remarkable, though. It’s the eerily similar shape of that value. He has recorded between 17 and 23 outs in every start. He walked no one in one start; one batter four times; two batters eight times; and three batters five times. He’s allowed one hit once, two hits once, and between three and seven hits in the other 16 outings. Beginning with the most recent, his last nine starts have seen the following strikeout tallies: 5 5 6 5 4 4 6 6 5. His seasonal ERA hasn’t run higher than 2.85 after any start, and is now 2.69.
This is all neat, but not necessarily meaningful. Unless you’re a very aged baseball fan with an exceptional memory and systematic mind, you probably don’t have a sense for what a six-inning, five-strikeout, two-walk, four-hit, two-run outing is worth. You probably struggle, as I do, to decide whether it’s really better to dominate so thoroughly in four games that the team can’t help but win, then struggle and put your team behind the 8-ball three times, or take seven steady starts like those.
Happily, we have a system for measuring this, albeit imperfectly. Baseball-Reference tabulates a stat called RE24. Basically, we know the run expectancies (RE) of each of the 24 base-out states within an inning (no runners on, no outs; runner on first, two outs; etc.), and so this metric simply subtracts the run value of the outcome of each plate appearance from the RE of the base-out state in which it began. By that measure, Wood has had 15 starts that made positive run contributions out of 18 total outings, and of the three that hurt, two were by less than one run, on average.
Win Probability Added (WPA) is another framework, based this time on the chances of winning the game, and how each plate appearance by opponents impacts those odds. Five Wood starts come in in the red in this metric, but the margins are slim again: He hurt the team’s chances to win by just 1.6 percent, 8.4 percent, 9.2 percent and 10.1 percent in four of those. Twelve of his starts have helped the cause by 11.3 percent or more.
So it seems like Wood’s consistency is a huge positive asset. Even his bad starts have been the kind of thing a big hit or two can easily overcome. A pitcher who goes six innings and lowers the chances of winning the game by 10 percent has by no means hamstrung his teammates. An average outing really lets the game breathe, and the rest of the team has a chance to pick up the slack if needed.
Of course, we’re looking only at results. There are things in Wood’s profile that throw up some caution flags. He’s an extreme fly-ball pitcher, and the reason those guys are usually not as consistent as others is that if a ball flies out of the park at the wrong moment, the tone of the outing changes quickly. Indeed, of Wood’s 26 starts last season, five were as good or better than his best this year, but six were also worse than anything he has done in 2013. In those six nightmare starts, Wood allowed a total of 18 home runs. This season, Wood has only allowed as many as two home runs in one outing, and both were solo shots, at that. Wood pitches to poor contact and gets a lot of infield pop-ups and shallow flies, but it’s probably a safe bet that the homer bug will bite him a few times from here onward, and that his marvelous consistency and excellence will erode somewhat accordingly. It’s also a safe bet, despite that effort to induce pop-ups and weakly hit balls, that Wood’s league-low (for more than 50 innings) batting average against on balls in play (.215) will rise. He’s not quite this good, is what I’m saying.
Still and all, there are things Wood is doing really well, things that have made him both more effective and more reliable than he was in Cincinnati (before the Cubs dealt for him prior to 2012) or in Chicago last season. For one thing, he’s missing more bats. They aren’t turning into strikeouts, per se, but his frequency of foul balls has shrunk substantially.
That means he’s having fewer pitches fought off and is able to be more efficient. His rates of 3.80 pitches per PA and 14.9 per inning are career-best marks, so he’s able to pitch deeper into games. The extra out or two per outing he’s getting are valuable, because usually, those are the outs for which the fourth- or fifth-most effective reliever in the bullpen would be responsible.
Wood’s command has also improved, although not by leaps or bounds. His walk rate is down from 8.3 percent to 7.6 percent, and yet, it seems Wood is less willing to give in when behind in the count. That and a strong Cubs defense are part of the reason for his outrageously low opponents’ BABIP, so it’s not all luck.
The Cubs have an asset here. Wood is under team control for three more years after 2013. It’s increasingly clear that a fringy fifth starter or lefty relief option has turned into a certain commodity. Wood can’t pitch at the front of a contending rotation, but as the fourth guy in a strong corps, he will be around when the Cubs are good again.
Pitcher Development Taking Big Strides
There are two crucial elements to building a winner in MLB.
A team has to identify and acquire talented, strong players. They have to be good at scouting, evaluation, negotiation and follow-up. One has to be on the lookout for talent that will play in the big leagues. This is the first part, and the hardest.
The other thing a good team has to do, though, is get the most out of the players they do acquire. The balance between scouting and player-development is not clear. Some think talent will always shine through. Others believe more in the ability of good organizations to milk extra value out of moderate or meager talents.
I tend to fall more into the latter camp. Scouting is still the most important thing, but I don’t believe it to be the only thing. In the latter years of the Jim Hendry era, the Cubs were very bad at player development, and in particular, at pitcher development. The pitching prospects they traded or gave up on did better than the ones they brought along and installed as big-leaguers, and their big-league hurlers often seemed frustratingly unable to tap fully into their talent.
That’s changing. In Year One of the Epstein-Hoyer regime, the Cubs got a tremendous run out of Ryan Dempster, maximizing his trade value. Ditto Paul Maholm. The front office traded Andrew Cashner, who has since been inconsistent and oft-injured, and gave a full-time starting gig to Jeff Samardzija, who (with the help of the instructional staff) turned a major corner and announced himself as a solid mid-rotation power pitcher. They added Wood in trade, but also drafted tons of arms after the first round of the June 2012 draft, then invested in some injury-risk/upside arms around the trade deadline—guys like Arodys Vizcaino and Barret Loux.
They made more minor additions over the winter, but perhaps the biggest augmentation they made to their pitching pipeline came in October last year, when they lured long-time Vanderbilt pitching coach Derek Johnson into the fold as the minor-league pitching coordinator. Johnson was as well-reputed as any college coach in years. He oversaw the development of David Price and Mike Minor, among others, during his time there. Adding Johnson was a declaration of intent, to build better pitching staffs from within. The front office clearly liked what it saw in the progress of Wood and Samardzija, Dempster and Maholm, and doubled down on pitching as a skill they could cultivate, not just buy at maturity.
Scott Feldman was one of their targets, a guy they brought in over the winter mostly in hopes of getting something good out of him, then getting something good for him. That plan worked to a tee. They ended up deadling Feldman for Jake Arrieta, a failed top prospect from the Orioles organization (a franchise still failing as spectacularly to turn talent into mound production as the Cubs had been before the change), and are now asking Johnson (among others) to work more magic. Arrieta will spend some time in Triple-A making the adjustments the Cubs will want to see, as Wood did just last season. By the end of this year, though, I’m betting Arrieta will be in the big-league rotation and showing signs of life.
Samardzija, who uses a similar repertoire, struggled (like Arrieta) with command as a younger man, and blossomed beyond age 25 himself, offers a blueprint for resurrecting the new arrival. Minor-league arm Matt Loosen gave reason for optimism about that process this week, too, when he threw a nine-inning no-hitter in the Florida State League. Loosen has tapped into some good raw stuff this year by improving his line to the plate, a key part of Johnson’s pitching philosophy. Johnson is a guru of mechanics, emphasizing staying tall and closed, and taking one’s nose toward the catcher’s mitt at release. Loosen has really responded to that instruction, and is back on the prospect radar. Arrieta is a candidate for a similar transformation. With a flood of new college arms nabbed in this year’s draftand some pricey international amateur signings coming aboard, the Cubs have added a great deal of talent, and their development system for them seems much stronger than it was a few years ago.
Next post: The Locke Impediment, The Rasmus-Jay Gambit, and Other Trade Deadline Dilemmas
Previous post: The Limp and Holey Hosiery, and What Must Be Done