Each year around Christmas, the Hardywood Brewery releases their iconic Gingerbread Stout to their membership club in a box featuring eleven delicious variants. My wife and I, ever the beer dorks, belong to the club and my favorite variant each year is inevitably the Apple Brandy GBS, their traditional Gingerbread Stout aged in apple brandy barrels. It’s boozy yet smooth, somehow subtly marrying the ginger and honey of the original GBS with delicious flavors like cinnamon and baked apple. In 2018, we purchased several additional bottles and had one last bottle left just before Christmas 2019. We cracked it open, bracing ourselves for the delights contained therein... and it was flat. It was so flat, in fact, that it ruined the bottle, one of the rare handful of bottles of Hardywood products that disappointed me. That brings me to today’s topic- another beloved, aged product associated with barrels and a steady track record that suddenly went flat. Today we’re going to talk about Matt Carpenter’s 2019, and take a look at some historical hitters who have had a similar path. We’ll gauge what a rebound in 2020 would look like for the Galveston Grinder.
You’ve seen this kind of work before, including last off-season. Since Dan Szymborski’s ZIPS comp was Doug Rader from the 70s, I’m going to cast a deep net using all non-pitchers since 1960. First, let’s establish some defining characteristics of Carpenter’s 2019. His geometric mean wRC+ in the two previous years was 131.76, and his wRC+ in 2019 fell to 95. That’s a difference of 36.76, and we’ll want players who experienced a similar drop. We’ll use all players whose decline was between 31.76 and 41.76. Carpenter was 33 years old last season, and age is going to be a factor. We want hitters that are at least 32 in their decline season. On the other hand, we don’t want ancient players, so we can eliminate any hitter over 35. Carpenter had 492 plate appearances last year, not quite full time but close to it. We’ll eliminate any hitter with less than 350 plate appearances in their decline season.
Doing all of that gets us a sample of 110 players, with Carpenter in the middle. His 2019 decline was driven in large part by a decrease in power. Let’s introduce that component. First, the geometric mean of his ISO+ in 2017-2018 was 139.27 and it fell to 89, a drop of 50.27. Our group will be between 40.27 and 60.27 points of ISO+ decline. Suddenly we’re down to 34 players.
Our last step will be to find hitters who walk a lot, as Carpenter does. Even in his down year, his BB%+ was 145. Our group will include players between 130 and 160. With that final cut, here are our Carpenter comps:
Matt Carpenter Comps
We have some issues right away. Greg Luzinski and Adam LaRoche retired after their decline seasons, and injuries limited Darren Daulton to five games in his rebound season. To compensate, we’ll plug in Doug Rader’s 1975 decline season (who am I to argue with Szymborski?). I would also add a player who just missed the cut with a lower walk rate- 2012 Kevin Youkilis and his decline season- except he also retired afterward. The next best we can do is David Ortiz and his 2008 decline.
Here is the progression of our eight comps, showing their two previous years geometric mean, their decline year, and their bounceback season for ISO+, BB+, and wRC+.
The V shape is what you want here. What you don’t want is a line that ends up lower than it started. Seven of eight enjoyed a modest rebound in wRC+. The non-rebounding outlier was Big Papi of all people, who roared back just one season later and didn’t really fade again for the last six years of his career. Walk rates also rebounded, and usually even surpassed pre-decline levels for several of these players. Yet again, Ortiz is the outlier who had two years of slippage. Those are great signs for Matt Carpenter in 2020.
ISO is murkier and generally not promising. Jayson Werth, Doug Rader, and Ortiz saw their power slip even further. Brian Giles didn’t recover anything and Helton’s rebound was very modest. Rocky Colavito inexplicably reached new heights following his collapse, while Ken Singleton and Pedro Guerrero split the difference between their previous level and their decline seasons. Guerrero’s bounceback even happened in his first full season in St. Louis.
If we calculate the harmonic mean of the rebounds, the Carpenter comps improved:
- 13.7% in wRC+
- 8.56% in ISO+
- 15.35% in BB%+
If Carpenter improves in the same way, his baseline for 2020 would be a 108 wRC+, 97 ISO+, and 167 BB%+. Compared to his 2017-2018 peak, those numbers aren’t thrilling but getting some of that production back into the lineup after last season would be a welcomed addition.
Complicating matters is the fact that Carpenter’s advanced data via Baseball Savant went in some bad directions last year. His barrel percentage, launch angle, and exit velocity all declined significantly. No amount of positive historical comps can overcome declining skills if that’s what ailed Carpenter last year. Admittedly, we have no clue if Rocky Colavito was getting good loft in 1967 and 1968. Perhaps our control group had similar issues to Carpenter, but there’s obviously no way to know. On the other hand, there are signs everywhere that Carpenter was trying to play through nagging injuries that sapped him of strength, so much so that the Cardinals have talked about the strength program they’ve given to Carpenter for the off-season. If his problems last year were injuries rather than age, that’s encouraging that he’ll be able to reach the upper end of his comps here.
Moreover, the game is changing at breakneck speed. If we had looked at historical comps for players like Max Muncy, Jose Ramirez, or Justin Turner in recent years, the comps would have been significantly less productive than what that trio became. Hitters today have more technological tools than ever before to improve, and no player is going to work harder to correct their problems than Matt Carpenter. A bounceback to that 108 wRC+, 97 ISO+, 167 BB%+ level seems likely, but even that undershoots what ZIPS projects for him. In short, the error bars are huge and there’s probably a lot more value still left in his bat.
By the way, this year’s apple brandy GBS variant was a supercharged imperial stout- the Apple Brandy Mega GBS. It was amazing, yet another historical comp for a Matt Carpenter rebound.