Yovani Gallardo is in decline.
As our J.P. Breen explained last month, Gallardo’s problems have compounded with his increasing inability to throw his curveball for strikes. Without the curveball to worry about, hitters sit back and wait for Gallardo’s fastball. Unfortunately, Gallardo’s fastball is also slowing down, meaning that batters are locking in on a pitch that is also getting easier for them to hit. Gallardo’s strikeouts have plunged, and his overall production is trending the same way.
Jeff Sullivan claims that pitchers like Gallardo usually do not bounce back once a decline like this begins. Commenting on Sullivan’s article, Mitchell Lichtman felt that while Gallardo’s strikeout numbers were likely to rebound, “we can expect that his days of striking out 25% of his opponents are gone.”
It’s possible that Gallardo will rebound satisfactorily and function as a useful mid-rotation starter in 2014. But it is very likely that he won’t. If so, the Brewers must find some other way to use Gallardo effectively.
My solution to the Brewers’ Gallardo problem, at least for 2014, is simple: it is called “Tom Gorzelanny.”
To understand why, we start with something fundamental: the basic difference between starting pitchers and relief pitchers. In general, Starters are expected to last about three times through the opponent’s batting order (~100 pitches, or 6 innings), while Relievers generally are expected to last only once through the batting order, if that. Although people often distinguish between Starters and Relievers by discussing fastball velocity, pitch variety, and physical makeup, these characteristics are ultimately a function of this Time Through the Order (TTTO) profile. Provided that a pitcher can meet the TTTO expectation for his role, it is largely up to him how he does it.
This TTTO tendency is clearly on display at Baseball Reference, which tracks the major-league-average performance of Starter and Relief pitchers each time through the order:
|2013 Season Situation||Pitcher Effectiveness|
|1st PA in G as Starter||96|
|2nd PA in G as Starter||104|
|3rd PA in G as Starter||112|
|1st PA in G as Reliever||93|
|2nd PA in G as Reliever||109|
|3rd+ PA in G as Reliever||148|
For ease of reading, I renamed Baseball Reference’s “tOPS+” as “Pitcher Effectiveness.” “100” is the league-average performance for a batter during a plate appearance. When the number is higher than 100, the batter is performing better than the pitcher. When the number is lower than 100, the pitcher has the advantage. The further away from 100 we get, the percentages skew even further in favor of the batter or pitcher.
As you can see, in 2013, starters were, on average, able to hold batters to 4% below-average production during their first time through the order (the first inning or two). Having already seen the starter once, batters proceeded to produce 4% over average the second time through the order (typically the third or fourth inning) — a gain of 8 total points — and by the third time through the order (the fifth or sixth inning, usually), batters were hitting the typical starter 12% better than average — a total gain of 16 percentage points from the beginning of the game. In other words, there’s a reason scoring tends to tick up in the middle innings of a game, as starters tire and managers guess poorly when it is time to pull them. (Elite starters often don’t get through the batting order for a third time until the seventh or even eighth innings, which is why they can get deeper into games.)
Relievers, on the other hand, tend to be more effective than starters in their first time through the order, but much less effective when forced to keep pitching. Because they know they will not pitch for long, relievers can throw harder than starters do, and thereby create sharper contrasts with their off-speed pitches. Tired hitters, especially late in the game, can have difficulty keeping up. But when forced to go multiple innings, relievers are at a definite disadvantage: as you can see above, batters have a 9% advantage over relievers the second time through the order, as opposed to the 4% advantage they have over starters in this situation. It doesn’t happen often, and the sample size is very small, but you can also see above what happens when relievers are forced to pitch a third time through the batting order, and it is not pretty.
So, what does this have to do with Yovani Gallardo? Well, everything really, because Yovani’s decline in 2013 was not as complete as you might think. There are certainly some starters who “lose it” entirely, whether due to injury or some other factor. But other starters retain much of their effectiveness, just not enough to sustain three times through the order. That may be what is happening to Gallardo. Witness:
|Situation||League Average Effectiveness, 2013 (above)||Gallardo, 2013|
|1st PA in G as Starter||96||90|
|2nd PA in G as Starter||104||96|
|3rd PA in G as Starter||112||119|
Despite his challenges, Gallardo’s problem in 2013 wasn’t that he was a terrible pitcher. In fact, he was solidly above-average his first two times through the opponent’s batting order, and downright excellent the first time through. Rather, Gallardo struggled once he got into the middle innings of the game, as the opponent’s lineup got one final crack at him. This is actually what we would expect, given the scouting report: with fewer pitches to worry about, combined with two previous looks, batters were able to key in on Gallardo and start hitting him — hard. Gallardo has certainly struggled with hitters late in his starts before; what’s important for our purposes, though, is that he can still be effective the first two times through the order.
So, what do you do with a starter who can pitch twice through the opponent’s batting order, but struggles getting through the third time? Well, you can find another pitcher on the roster with a similar problem, and set him up as a platoon partner. It so happens that the Brewers have the perfect candidate in Tom Gorzelanny.
Gorzelanny is a former starter who pitched adequately for other teams, but was moved to the bullpen after he was not able to sustain his effectiveness. However, as Brewers fans saw this year, Gorzelanny still has the pitch mix to compete a few times through the opponent’s lineup. His problem is the same as Gallardo’s: while his stuff is decent, it is not currently good enough to last that third time through the batting order:
|Situation||2013 Gorzelanny Effectiveness|
|1st PA in G as Starter||97|
|2nd PA in G as Starter||76|
|3rd PA in G as Starter||248|
|1st PA in G as Reliever||70|
Gorzelanny, while lethal as a reliever in 2013, was also acceptable the first time through the order as a starter, even better the second time through the order, but had no business whatsoever being in the game any further. That’s ok, because this is exactly what we are looking for: somebody who can last four to five innings and provide decent run prevention, which is what Gorzelanny does.
So there you have it: the Brewers could plan to use Gallardo for the first five or so innings of the game, and then have Gorzelanny “rescue” him and take over from there. Properly done, Gallardo and Gorzelanny should finish out most games without further assistance from the bullpen. Admittedly, making Gorzelanny a platoon partner for Gallardo would have other ripple effects that need to be considered: it would create (at least) a 5.5 man rotation, and may necessitate keeping an extra reliever. On the other hand, the Brewers would have a ready-made long man — as opposed to a mop-up “spot starter” — stretched out and ready to go whenever injury inevitably strikes somebody in the rotation. And Gallardo might reclaim some of that last velocity by throwing a bit harder, knowing that the team doesn’t expect him to go beyond five innings anyway.
Do I expect the Brewers to do something this unconventional, even if they should? Nope. Gallardo would probably balk at the idea, perhaps literally. The Brewers also desperately want their home-grown starter to regain his form, and further want to maximize his value if they decide to get rid of him. Platooning Gallardo is not going to achieve either of those goals in the short term.
But Gallardo’s current performance decline is not going to achieve those goals either. Moreover, it is possible that a lighter workload to start the season, after several years of full usage, could improve Gallardo’s performance. Over his career, Gallardo has tended to start slow, improve into the All Star Break, and then decline as the season progressed to its conclusion. A few months of five-inning work could keep Gallardo fresh, perhaps allowing him to retake the full starter’s role down the stretch. That sort of improvement could certainly interest potential trade partners, and would ultimately benefit the Brewers if they fulfill their dream of being a fringe contender this year.
It’s time for a Plan B for Yovani Gallardo. If the Brewers have a better one, let’s hear it.
Follow Jonathan on Twitter @bachlaw.