There’s been much consternation among Cardinal fans about the offense this off-season. The logic is fairly simple and it’s prevalent in so many articles about the team written over the last several months. “The offense struggled last year and looked awful in the NLCS. Now they lost Marcell Ozuna, their cleanup hitter. How will they improve in 2020?” It’s time to put some of this to rest, while still not letting the team off the hook.
FanGraphs has a great one-shot metric to describe a player’s offensive contributions. It’s aptly titled “Offense” and it’s the offensive component of their WAR calculation. Here’s what we’re talking about specifically:
Offense (Off) is a statistic that combines a position player’s total context-neutral value at the plate and on the bases. Off is a combination of our park adjusted batting runs above average and our base running runs above average and credits a player for the quality and quantity of their total offensive performance during a given period of time.
One important note- Offense measures runs against league average, and not against replacement level. Keep that in mind as we proceed. The topic du jour is replacing Ozuna, particularly at the plate. In 2019, Ozuna’s Offense was 8.8. Marcell Ozuna’s offensive contributions were 8.8 runs above league average. If we use the handy FanGraphs chart on the glossary page, we can see that he’s right in between their designation of Average (0) and Above Average (15). If you want some context, Ozuna’s Off ranks 85th out of 169 players with 450 plate appearances. That’s a perfectly fine and decidedly unspectacular output, and that’s what the Cardinals need to replace at the plate simply to replicate 2019.
Now let’s look at what the Cardinals have in house. To prove the larger point here, you’ll see their average Off in 2017-2018 next to their 2019 Off, and the difference between the two:
Cardinals Offense (Off), Pvs. 2 Yr. Avg vs. 2019
Player | 2017-2018 Avg | 2019 | Diff |
---|---|---|---|
Player | 2017-2018 Avg | 2019 | Diff |
Carpenter | 27.00 | -2.8 | -29.80 |
Goldschmidt | 40.20 | 17.3 | -22.90 |
DeJong | 7.65 | 2.2 | -5.45 |
Molina | -4.70 | -11.4 | -6.70 |
O'Neill | 5* | -1.7 | * |
Bader | 4.70 | -9.3 | -14.00 |
Fowler | 1.60 | 1.6 | 0.00 |
Wong | 3.85 | 12.9 | 9.05 |
Tyler O’Neill is tricky, as his previous two years average only includes his partial 2018 season, ergo the asterisks. Beyond O’Neill, what you see there is a lot of collapse, and the magnitude of the collapse can rightly be called unexpected. Add it all up and those seven players (O’Neill is excluded) amount to just under 70 runs of offensive contribution lost from their 2017-2018 average, even accounting for Wong’s significant step forward.
Because of the youth of Paul DeJong and Harrison Bader, for instance, that type of pronounced decline is abnormal. You can certainly expect some age-related decline for players like Matt Carpenter and Paul Goldschmidt, but not as dramatic as what happened. Because of Yadier Molina’s advanced age and position, there’s an argument to be made that his decline was legitimate. In fairness, this is all overly simplistic. For example, you’d want to weight 2018 more than 2017 in the two-year average, and you’d absolutely need to include an aging curve. Nor do I mean to imply that they’re sitting on a 70 run powder keg of rebound. Still, 69.8 runs lost from the previous two years from seven players is massive. At least some of it should come back.
Let’s return to Ozuna. His departure creates a deficit of 8.8 runs above average at the plate. Go back and look at that table. It’s reasonable to think those seven players can produce 8.8 more runs. If Carpenter recovers only a third of his production lost- to say nothing of rebounding completely, just a third of the way- that alone would more than cover Ozuna’s lost production. A third of a rebound for Goldschmidt would be 7.6 runs, almost covering the loss of Ozuna. Harrison Bader performing closer to his expected stats instead of the dumb luck he had in 2019 would more than compensate for Ozuna. Maybe any four of them improve by a scant two to three runs, collectively adding up to Ozuna’s lost output. There are a lot of roads to replicating the lost runs.
This all also assumes that O’Neill, Lane Thomas, Justin Williams, and Dylan Carlson collectively land at 0.0 offensive runs contributed. Given their inexperience, that’s a perfectly rational assumption. Again, though, it’s not unreasonable to think that a quartet featuring a current top 10 prospect, a recent top 100 prospect, and... well, Lane Thomas and Justin Williams- might exceed 0.0. You certainly wouldn’t assume they’d replicate Ozuna’s output, but the goal is to hold down the fort enough at the plate that reasonable, modest rebounds from Goldschmidt, DeJong, Carpenter, or Bader make up the difference. On top of that, the quartet can surely add value with the glove and on the bases that Ozuna did not, further softening the blow of Ozuna’s lost bat.
Here’s the problem. So far, we’ve only discussed replacing Ozuna. That’s a fine goal, but it means the Cardinals would be 14th again in Off (pitchers not included) and wRC+, and 15th in wOBA. In other words, they’d have another average offense. They need more than that. Even if any one of Goldschmidt, Carpenter, Bader, DeJong, et al. recover a third or even half of their lost production, it only replenishes lost value instead of pushing the team forward. If the goal is to replace Marcell Ozuna’s production, the simplest way to do that would have been by signing Actual Marcell Ozuna™. Not doing so leaves them vulnerable if players like Kolten Wong, Tommy Edman, or Dexter Fowler decline regardless of rebound from the others.
All of which is to say that the ease of replacing Ozuna with in-house options does not let the Cardinals off the hook. They can replace him in-house, but this wasn’t a binary situation. They could have had both internal improvements and the consistency that Ozuna provides. It’s fine if they didn’t want to do that and I certainly understand not wanting to make that commitment, but there was no other alternative plan. Now they’re more prone to variance than they should be.