Steamer and I: Yoenis Cespedes – A Review

Onward and forward we move with the Steamer and I series recaps, pitting my Pod Projection against Steamer! Today, I’ll review how we forecasted Yoenis Cespedes. Unlike Mike Trout and Yasiel Puig who I previously recapped, I was actually significantly more bullish than Steamer on Cespedes. Let’s find out exactly what each system was projecting versus how Cespedes actually fared.

For the projection comparison, all 2015 and Steamer counting stats have been extrapolated to the same number of plate appearances I forecasted. I also decided to add a line for his 2016 counting stats extrapolated over the plate appearance projection, given that he missed our playing time forecasts due to injury.

Immediately you’re reminded how sizable the gap was between the wOBA projections from Steamer versus Pod. For an established veteran, a 0.020 difference is huge. Although Steamer forecasted a precipitous drop in doubles rate, something I initially questioned, it was nearly right on the money when comparing the extrapolated total. That represented an enormous decline from his 2015 level. His speed also completely disappeared, as he hit just one triple, after finishing with six over the previous two seasons. Obviously, both Steamer and I were off on that forecast.

The home runs, and overall power output, is one area where Steamer and I disagreed. I put more stock in Cespedes’ 2015 home run barrage than Steamer did and figured that any regression in HR/FB rate would be countered by a rebound in fly ball rate, which had tumbled in 2015. Turns out, although I was far closer on the ISO than Steamer, we were both low. Cespedes actually matched his 2015 ISO, and boosted his previous career high HR/FB rate set in 2015 even further. The FB% also increased, as I suspected it would, though still fell far short of where he sat in 2013 and 2014. Interestingly, his fly ball exit velocity and Barrles/Batted Ball Event both fellow from 2015 to 2016, suggesting that HR/FB rate jump was a total fluke. Just another head scratching driver of the leaguewide power surge.

Cespedes was never much of a walker, as he has always swung more frequently than league average, mostly fueled by lots of swings at pitches outside the zone. Steamer and I saw a slight rebound in walk rate this year, but generally in his typical range. Instead, he nearly doubled his 2015 walk rate and posted a new career high. He cut down on his Swing% and O-Swing%, both to levels that set new career lows, pushing his Swing% down to only marginally above the league average. Such a change surely should have increased his walk rate, but it seemed like it moved by a much greater degree than it should have. Some of the explanation is provided by a career high rate of intentional walks, which is not going to affect his Plate Discipline metrics.

Steamer and I took different paths for Cespedes’ 2016 strikeout rate projection. Steamer projected his strikeout rate to jump, while I expected it to improve slightly from his 2015 mark. I was more accurate and predicted the correct direction, as his strikeout rate actually improved substantially, dipping below 20%. He got his SwStk% to about league average, and he may have struck out even less if that wasn’t paired with a drop in Swing%.

Steamers .292 BABIP projection seemed overly pessimistic, as it would have ranked as the second lowest mark of his career, though not by much. I projected some regression from Cespedes’ 2015 mark to .300, which was right around his career average, and he came in just below at .298. He hits a lot of fly balls and pop-ups and doesn’t go the opposite way very often, but obviously, hits the ball hard. He’s never going to be a big BABIP guy so he should continue to bounce around league average territory.

So overall, Cespedes actually increased his wOBA ever so slightly from his big 2015 performance and set a new career best mark. Since offense league wide also increased, his wRC+ actually declined by a point, but still, most didn’t expect him to sustain that breakout and he did. Unfortunately, given his age (31) and pessimistic Statcast metrics, I’m thinking he’s now going to be overvalued in 2017 drafts.