Chad Bettis needs to stay in the starting rotation

After his first two seasons in the big leagues brought nothing but despair—69⅓ innings pitched, 97 hits allowed (12.6 hits per nine innings), a 6.88 ERA, 3.9 walks per nine innings, and just 5.6 strikeouts per nine innings—Chad Bettis appeared to turn the corner in his 2015 season. The then 26-year-old put up career best marks in ERA and strikeouts, walks, hits allowed, and home runs allowed per nine innings as he transitioned back to being a full-time starter and shifted his primary breaking pitch from a slider to a curveball.

Now that his promising 2015 season in the rear view mirror, 2016 Chad Bettis appears to be back to his old, read "bad," self. His ERA has spiked from 4.23—eight percent better than league average—to 5.55—14 percent worse than league average. His strikeout totals have also dropped from 7.67 per nine innings—better than the average starting pitchers’ 7.4 strikeouts per nine innings in 2015—to 7.04—much worse than the average 2016 starter’s 7.73 strikeouts per nine innings.

Pair this poor performance with Rockies’ top pitching prospect Jeff Hoffman appearing to be ready to join the Major League rotation, and you understandably get speculation that Bettis could be the odd man out if/when Hoffman arrives. Would that be the right move? As always, there’s more at play here than what we see on the surface. We’ll have to look deeper into the numbers to find out the answer.

Strikeouts are obviously important for any pitcher, but another critical component is limiting walks. For the Rockies, limiting walks may be even more important, and Bettis has actually gotten much better in that department in 2016. After walking 3.29 batters per nine innings in 2015, Bettis has dropped that number all the way down to 2.21 in 2016. That takes him from about half a walk per nine innings worse than league average to about half a walk better than league average. It doesn’t appear to be a fluke, either, as Bettis boasts career highs in percentage of first pitch strikes thrown as well as the overall percentage of his pitches that have been strikes. The reduced walks have led to Bettis having an expected ERA nearly identical to what he had in 2015:

If we’re to trust these numbers, Bettis has been slightly better than the league average starter so far in 2016, just like he was the year before. If that’s the case, why have the surface numbers for Bettis been so much worse? One possibility is simple bad luck. Let’s look at some categories that have at least some element of luck involved in them like batting average on balls in play against (BABIP), home run rate (HR/FB%), strand rate (LOB%), and the amount of infield (IFH%) and bunt (BUH%) hits allowed.

Well, it looks like we may have our answer here. Bettis appears to have been less lucky across the board, which matches up with the idea we got from the ERA predictors that he hasn’t actually been any worse this season. These aren’t perfect statistics, though, and maybe they aren’t accounting for the quality of contact Bettis has allowed this season compared to last. There’s only one way to find out!

That looks...basically the same as it did last year. There’s a marginal uptick in the amount of hard contact allowed, but an even bigger marginal uptick in the amount of soft contact induced. Certainly nothing here suggests Bettis has gotten hit hard enough to cause the jump we’ve seen in things like his BABIP and home run allowed rate. Another thing that immediately comes to mind when we see this happen to a pitcher is a dip in fastball velocity, but that isn’t the case with Bettis either, as he is averaging the exact same 92 mph he averaged in 2015.

Maybe everything isn’t so bad here. The surface numbers for Bettis absolutely look worse than they did a year ago, but everything else tells us he’s just as good as before. That doesn’t mean that he should be expected to carry the Rockies’ rotation, but removing him from it entirely might be just as big of a mistake.