Carl Pavano and the inside sinkerby Harry Pavlidis
June 23, 2009
Carl Pavano made his name on the disabled list in New York. Now he's pitching for the American League's worst team, but having a better go at it. Pavano hit the free agent market after the 2004 season, and got the red carpet treatment after an 18-win season.
Pavano's 2004 ERA of 3.00 wasn't an accident, but was a quarter of a run below his xFIP. In infamously limited time over the next four years, Pavano posted xFIPs of 4.42 in 2005, nothing in 2006, 4.35 in 2007 and 5.47 in 2008. Sure, those last two seasons included just nine starts for Pavano, but I'm not sure how many people expected an improvement in 2009 that, to date, has resulted in an xFIP of 4.12 (along with a "real" ERA of 4.18).
To be honest, I'm not 100 percent sure. But Pavano's sinker may hold the clue.
Pavano's sinker is his favorite pitch, and has been used on 42 percent of his pitches in 2009. That's a small increase over his 2008 percentage, which was accumulated over just seven games. It's not much to go on, but some things have changed with Pavano's sinker: namely its location and result.
During Pavano's short run in 2008, his sinker yielded a rv100 of 2.38, which means, over the course of 100 pitches, his sinker would give up 2.38 more runs than the average pitch. In 2009, that number hasn't become a sparkler, but has crept closer to zero, currently at 0.86.
rv100 is a rate stat, so, to put it in context, in just seven starts in 2008, Pavano allowed five more runs than an average pitch on his sinker alone, approaching a run a start (or run value above average, RVAA, to be accurate). In 2009, over 14 starts, the sinker has yielded an RVAA of 4.3. Seven games, five runs last year; 14 games, four runs this year. That's a lot better. So far, he's saved himself a good handful of runs relative to 2008 on his sinker alone.
Looking at Pavano's pitch locations (I'll spare you the gory details), I found the sinker has been placed on the inner part of the plate, or just off it, 25 percent of the time, compared to 18 percent in 2008. Along with that shift inside, Pavano has fared better with the sinker in that location. This move inside is reflected in a drop of pitches on the opposite side of the plate, and even more so on pitches wide of the plate.
In 2008, nearly 25 percent of Pavano's sinkers missed outside. This year, just 15 percent have missed wide. He's also throwing more sinkers over the middle of the plate, at no greater cost, rate-wise.
So, he's throwing the sinker inside more, and having more success. On pitches in off the plate, Pavano's rv100 has actually crept up about a run, but, on the pitches that stayed on the inner part, or just off it, Pavano's rv100 has gone from 12.12 in 2008 (yes, that's awful) to -2.08 in 2009, which is very good.
We're talking about one pitch in one location, but it reflects a change in approach along with a change in results.
This is a brief interlude from the park factor work that I had planned for this week. I'll resume that thread next time.
References and Resources
PITCHf/x data from MBLAM, classifications by the author.
Harry Pavlidis admits he has a baseball problem. He is the founder of Pitch Info LLC, His pitch classifications power the player cards at Brooksbaseball.net. Feedback, questions and comments are appreciated - Email firstname.lastname@example.org and Twitter @harrypav