Reviewing 2017 Pod Projections: Kyle Hendricks

It’s time to recap some of my 2017 Pod Projections! This preseason, I begun the series with one of 2016’s most surprising pitchers, Kyle Hendricks. We all figured that even backed by the historically strong Cubs defense, he was quite a bit fortunate en route to a sub-3.00 ERA. But how much regression was I projecting and how did that compare to his actual results? Let’s find out.

IP: 188 (31 games started) Projected vs 139.2 (24) Actual

Hendricks missed about a month and a half with a hand injury, limiting him to just 24 starts. His IP/GS declined from 6.3 in 2016 to just 5.8 this season, which is pretty significant. Since the 2016 mark was achieved with some help by run preventing abilities that he’ll probably never reach again, I wouldn’t expect a rebound to 6.3 IP/GS anytime soon.

K%: 21.8% Projected vs 21.6% Actual

Though his fastball velocity dipped a full two miles per hour to just below 86 and he generates a below average SwStk%, he still managed to strike out batters at a league average clip. That’s thanks to a spectacular changeup. I had projected a decline in strikeout rate based on the poor velocity and assumption that the SwStk% of his changeup would be difficult to keep so insanely high, and those assumptions proved correct. He’ll probably remain in the 20% to 22% range for a while, but he cannot afford to lose any more fastball velocity.

BB%: 6.3% Projected vs 7.0% Actual

After remaining consistent having thrown strikes between 64.9% and 65.5% in each of his first three seasons, Hendricks’ control suddenly deserted him, as his strike percentage fell to 63.9%. That was the driving force behind the walk rate spike. It’s always a guess as to why a pitcher will suddenly lose some control for a season and then find it again the next, but I’d figure some rebound in 2018.

GB%/LD%/FB%: 49% / 20.5% / 30.5% Projected vs 50.1% / 20.8% / 29.1% Actual

Hendricks’ batted ball profile had remained rather consistent, as most pitchers’ do and he posted another season right in line with his career averages, which meant I was pretty darn close. The slightly higher grounder rate versus fly ball rate was a good thing, considering that inflated HR/FB rate. The fewer fly balls, the better!

HR/FB%: 10.5% Projected vs 14.8% Actual

I couldn’t believe Hendricks owned that elusive HR/FB suppression skill, especially in a park that favored home run power. So I projected an increased mark toward what had used to be a league average HR/FB rate mark. Of course, I had no idea the league was going to go home run bonkers and Hendricks ended up being one of the victims, erasing all hopes that he was an outlier. I wonder how much that terrible fastball velocity had to do with the surge.

BABIP: .280 Projected vs .281 Actual

All the super advanced metrics confirmed that Hendricks did generate soft contact and seemingly deserved a suppressed BABIP in 2016. But for one, suppressed BABIP doesn’t necessarily mean a .250 BABIP, and two, it doesn’t mean it’s actually a repeatable skill. I gave him some credit for the skill just in case, but still figured severe regression toward the league average and amazingly nearly hit the nail on the head. I’m still not sure how he does it given a ground ball heavy profile and nothing else that stands out, but a .274 BABIP over 590 innings means something.

Below is a comparison of all the preseason projections and the actual results, with highlights for the system with the closest forecast:

I didn’t bother to highlight IP or wins because of the injury. As I would have guessed, my projection and Steamer were closest on the most metrics. The Fans were most optimistic, as usual, and almost nailed the ERA. Interestingly, I’m typically an aggressive regressor like Steamer, but easily came in with the lowest BABIP projection.

Heading into 2018, I’m seriously concerned about that fastball velocity and Steamer clearly is too, projecting a sub-20% strikeout rate and 4.03 ERA. It depends on his price, of course, but I’d probably stay away.