I’ve written quite a bit this year about pitcher aging — specifically, trends in velocity loss for pitchers. There are two general findings that I want to revisit today and apply to pitchers from 2012; the predictive power of velocity loss in July and end of season velocity, and the impact of losing velocity in one season on next season’s velocity.
First, a pitcher’s velocity will tend to vary throughout the year. Trying to get a read on whether a pitcher is having trouble velocity-wise during a season is difficult if you simply compare to last year’s overall velocity. So I compared a pitcher’s velocity in each month to their velocity the previous year in that same month and found that pitchers who lose at least 1 mph of velocity in July are 13.7 times more likely to finish the entire year down at least 1 mph.
Second, 91% of pitchers that do finish a season down at least 1 mph compared to the previous season will lose additional velocity the following season (average decline of 1.6 mph), with only 7% regaining some (but, likely, not all) of that velocity back.
With the close of the 2012 season, I checked back on how well July-over-July velocity trends predicted full season declines as well as which pitchers ended the season losing over 1 mph off of their fastball.
Using PITCHf/x data, I calculated the average fastball velocity for all pitchers with at least 40 innings pitched this year and last year and that had the same role each year (i.e. pitchers needed to be either starters or relievers in both seasons to control for artificial velocity changes). For the fastball calculation, I included four-seam, two-seam, sinkers, and cutters. I then compared the average velocity for each pitcher, year-over-year.
Of those pitchers that were down at least 1 mph for the month of July compared to last July, 55% finished the season down at least 1 mph. Compare that to only 2% if they were not down at least 1 mph from last July. That means pitchers that were down at least 1 mph in July were 23 times as likely to finish the season down, more than–but in line with–the overall finding earlier this year (13.7 times as likely). If we just isolate to starting pitchers (>=150 IP in both years), the likelihood jumps to 42 times (86% versus 2%).
The amount of difference from July-to-July is also quite predictive of the year-to-year difference we see. Again, staying with just starting pitchers this season, the July-to-July difference in velocity explained about 70% of the variance in velocity difference, year-to-year (compared to ~60% in a larger study from 2003-2011). We would expect some relationship, since July velocity helps make up yearly velocity, but July is only one of six months that make up overall velocity. So, if you are looking for one, July velocity variation does appear to be a solid proxy for how a pitcher will finish the season.
Now that the season is over, here is the list of the 40 pitchers whose fastball velocity declined at least 1 mph from last season (I’ve included a few pitchers that lost -.9 just as a reference):
|Pitcher||FBv2012||FBv2011||Diff||FBvJuly2012||FBvJuly Diff||Role||Injury Type||Total Days Missed|
Role refers to whether the pitcher was a starter (S) or reliever (R) in both 2011 and 2012. I’ve also listed whether the pitcher sustained an injury in 2012 and how many total days they spent on the disabled list (courtesy of Jeff Zimmerman).
Looking at the list, Tim Hudson jumps out as one of the bigger concerns heading into next season. Hudson’s velocity has been fairly consistent, year-to-year, since 2008, never once increasing or decreasing by more than .8 mph. This year, however, the 36-year-old experienced a decline of 2.5 mph. And while Hudson did manage to post solid numbers this year (once again posting a sub-100 ERA- and FIP-), his strikeout rate declined almost 4.5% to 13.6%–his lowest total since 2004.
Hudson’s velocity was down consistently in each month in 2012 compared to the same time last season. However, Hudson did miss 26 days with a back injury in April and it’s possibly this injury lingered throughout the year. Either way, a velocity drop of that magnitude for a 36-year-old pitcher should not be ignored. Ninety-one percent of pitchers in their thirties that suffered at least a 1 mph velocity loss lost additional velocity the following year–a rate 1.5 times that of pitchers in their thirties that didn’t lose at least 1mph.
The Braves feature another front-line starter on the list; Tommy Hanson. 2012 represented Hanson’s second straight season where his fastball declined by more than 1.5 mph. Hanson declined to an average velocity of 91.1 in 2011 and all the way down to 89.6 this past season. That is not the trend one wants to see for a 25-year-old righty.
A number of other high-profile starters litter the list.
Earlier this year I argued that we are starting to see evidence of CC Sabathia’s decline, most notably because of the velocity drop he suffered this year. At that time, Sabathia’s velocity was down over 1 mph from the previous June. CC subsequently spent time on the disabled list for a hip strain from June 25 through July 17 and again for elbow soreness in August. Those trips could explain some of the velocity loss, however he’s was never tracking to his established velocity.
If you take Sabathia’s average monthly velocity from 2009-2011 and compare it to last year it’s clear his problems began well before June (note the y-axis is truncated):
Sabathia ended April with an average velocity of just under 92 mph. Compare that to an average April velocity of roughly 93 mph the previous three years. Time on the disabled list likely explains why his velocity never increased the way it general does as the season progresses (cresting in July), but now that CC is heading to see Dr. James Andrews about his elbow the question will be whether he needs surgery to correct what appears to be an existing issue–albeit one that had not impacted his velocity in this way before. The Yankees should be concerned. From 2009-2011, Sabathia generated -20.5 runs per 100 fastballs clocked at less than 92.5 mph. This year, Sabathia generated 21.9 runs per 100 fastballs clocked under 92.5–virtually all of his fastballs. Sabathia still put up solid numbers in 2012 (ERA- 81, 23.7% K%, 5.3% BB%), but his HR/FB rate jumped to 12.5%–highest in his career. Eighteen of his 22 home runs allowed came on fastballs (82%). The previous three years? Only 56%. All great pitchers must adjust to reduced velocity at some point, but those adjustments typically take time.
The Phillies’ Roy Halladay suffered a 1.8 mph drop during his age-35 season. Halladay started April down 1.8 mph (89.1) from the previous April (90.9), and while his velocity did increase going forward it was only by the slightest of margins. Before Halladay suffered a back strain that landed him on the disabled list at the end of May, his fastball averaged 89.5 mph. After he came off the disabled list in July his velocity averaged 89.3–essentially, no improvement. Like Hudson, Halladay is on the wrong side of 30. So, despite a phenomenal work ethic, it’s likely that Halladay’s velocity won’t simply bounce back with the benefit of an off-season.
Tim Lincecum and Josh Beckett are two pitchers I previously identified as at-risk pitchers when it came to velocity this year and both finished down over 1.5 mph from last year. Lincecum’s velocity has been declining for some time now, but this year the drop was greater than we’ve seen recently and didn’t appear injury-related.
Beckett, who turned 32 this past season, saw his fastball decline 1.6 mph and hovered around 91 mph the entire year. Since 2008, Beckett’s biggest drop was previously .8 mph between 2009 and 2010. That previous dropped could be tied to a back injury that cost Beckett significant time. In 2011, Beckett’s velocity declined again, but only by .5 mph. This year, Beckett again spent time on the DL, but for a shoulder issue. Between the shoulder injury, age, and the significant drop and it’s possible Beckett’s average velocity next year drops into the low 90s-upper 80s.
I could spend another 2000 words writing about each of the pitchers on the list, but then my editor would kill me. Suffice it to say that these kinds of velocity drops warrant serious attention. There are a number of things that can cause these drops and may explain them away as episodic and isolated to last season, but the overall relationship between losing this much velocity and future velocity loss appears pretty clear–one typically follows the other.
Bill leads Predictive Modeling and Data Science consulting at Gallup. In his free time, he writes for The Hardball Times, speaks about baseball research and analytics, has consulted for a Major League Baseball team, and has appeared on MLB Network's Clubhouse Confidential as well as several MLB-produced documentaries. He is also the creator of the baseballr package for the R programming language. Along with Jeff Zimmerman, he won the 2013 SABR Analytics Research Award for Contemporary Analysis. Follow him on Twitter @BillPetti.