Garza's Spike, And What is "a Place"

I've been reading a lot lately about Matt Garza's fantastic 2011 season, and that it sort of came out of nowhere. Which is a really bizarre concept to wrap your head around, really. Like, I can imagine Matt Garza himself having this sort of pensive moment when left to himself.  "How does something come out of nowhere?" He might ask. "What is nowhere to begin with? Is nowhere a place? How can there be a place in the universe that is not a place? How does something not come from a place? What a weird word, 'place'."

what is 'a place' anyway? what is anything anyway?

His ERA dropped 60 points, his GB rate shot up 10%, and his FIP dropped a spectacular amount-- almost a run and a half. Most of that FIP improvement can be attributed to an eye-popping spike in his K rate-- a full six percent! That really jumps out at you. A quick glance at his stats page, however, shows that the real spike occurred two years earlier in 2009, and that the spike, as it were, is mostly attributable to an off-year in 2010. But that realization came too late. The seed of curiosity had been planted and I had to know, for the life of me,  how often a 6% jump in K rate occurs, sort of, out of nowhere?

I first queried how often a pitcher jumped 6% from one year to the next in the past 10 years (with at least. 19 names showed up.

K-SPIKERS


Some of these names seem not-so-surprising. Curt Schilling's second renaissance, once he left the tar pit of loserdom that was Philadelphia, is well-known in recent baseball lore, so it's not exactly a shocker that he shows up here. And we can all remember the break-outs of Weaver, Verlander and Lester just a few seasons ago.And yet, conversely, some of these names may seem not-so-not-so-surprising. I'm looking at you, Esteban Loaiza.

But flukes or not, this is an interesting group of pitchers no doubt. Some of them, though, I suspected only made this list by virtue of having an off-year previous to the spike. So what looked to be a 6% jump was really only a 2 or 3% jump in the long run. And don't forget, that I am only seeking out strikeout surges that came out of nowhere. Looking to eliminate theses scenarios where the spike came out of somewhere, therefore, I then required all pitchers to have increased their strikeout rate by 6% in both the previous year (Y-1) and the the year before (Y-2).

LEGITIMATE NOWHERES AND FAKE OUTS 

*less than 150 IP in 'Y-2'

Only 7 of these pitchers, as it turns out, experienced 'legitimate' jumps, where their spike year was 6% higher than both their two previous seasons. Of the fake-outs (boxed in red), Blanton, Kile, and Peavy all experienced significant jumps (+4%) from their Y-2 season, just not to the degree that the remaining pitchers did. And Weaver, unfortunately, just misses the arbitrary cut-off at 5.35%. Both Traschel and Arroyo were coming off the worst seasons off their career, in so far as K/BB is concerned, and thus their spike years are, perhaps, deceiving. And Vazquez's spike, we find out, was simply a more dramatic version of the very typical schizo career patterns for which he is notorious.

(Bonderman was a rookie in his Y-1 season, did not have a Y-2 season at all and therefore fails to make this list.)

From these legitimate-nowhere K-jumpers then, which ones proved to have sustainable strikeout rates of that new magnitude beyond their spike season into the following year (Y+1) and the year after that (Y+2).

FLUKES AND STUDS

*less than 150 IP in Y+1, **less than 150 IP in Y+2, ***less than 150 IP in Y+1 and Y+2

 Of the remaining pitchers only 2 (count 'em 2) managed to sustain that +6% K spike in the subsequent season-- Schilling and Lester. But, interestingly, 4 of those pitchers who saw negative regression in their first post-spike year, regained that 6% in the next season (Y+2)!

Sheets and Myers both had the benefit of not playing full-seasons in that second year, so clearly they require asterisks. In their Y+3 seasons, both fell back below that 6% threshold. A.J. Burnett did sustain his success, if you are willing to disregard that he averaged barely over 100 innings per season for the next 8 years. In that time he sustained a strikeout rate 5.8% above his Y-1 showing, on average.

Matt Clement did in fact match his spike levels two years later, but never again came close to exhibiting that degree of dominance. He was out of baseball two years later. Similarly, Loaiza also bounced back to his spike-year heights after a year of regression, only to lose it for good the following season. He hung up his cleats 3 years later.

Lester and Verlander have done well for themselves in their time since, including a Cy Young and MVP season from the latter. Both, however, are getting along with a strikeout rate at around 4.6% higher than that Y-1 season.

Curt Schilling then, is the only pitcher to have sustained that 6% for 3 consecutive seasons. But, we need to remember that Schilling had already established himself as capable of this sort of dominance. He just had the odd case where he had two consecutive "off-years", casting the legitimacy of his spike into doubt.

That said...

To get an idea of how rare that is, let's query the entire century for pitchers who have jumped 6%, 'legitimately', and then held on to that 6% for 3 consecutive years. It's only happened 3 other times:



...