Rockies starters and times through the order

If you follow baseball, you likely know all about the times through the order penalty. The short version is that the more looks hitters get at the same pitcher, the better they hit ‘em. There are many reasons for this, of course, and we’re going to talk about some of them in this week’s Rockpile. The general approach here is simple: we’re going to take a look at some of the things that cause pitchers to get worse the more they face the same hitter and then we’re going to see how the big four members of the Rockies’ rotation (Márquez, Senzatela, Freeland and Gomber) fare in each phase. Also, if there’s a “fix” I see for some of these issues, we’ll go over that too.

Fatigue

The most obvious reason of them all. The more pitches a hurler throws, the more tired he’ll be (not always a linear thing, but still). Fatigue shows up in pitchers in many ways: velocity drops, loss of command, loss of movement, etc. For this piece, we’re mainly going to focus on velocity, because going over command would be opening up a whole new can of worms that I likely wouldn’t finish in time for a Rockpile. If I see some changes in pitch movement, I’ll show those as well.

Familiarity

Another fairly easy concept to understand. The more you see the same pitcher throw, the more used you get to his delivery, timing, movement, and so on. This is often represented by a decrease in strikeouts the deeper a starter gets into a game, among other things. We’ll look at drops in K%, and also Whiff% if I can find it.

Pitch Mix

This is related to familiarity, of course, but it’s kind of its own separate thing. As you’re probably aware, it’s considered bare minimum for a pitcher to have three proper offerings in order to start. Two-pitch starters really aren’t very common, and the ones who rely heavily on a two-pitch mix usually get punished severely by the times through the order penalty. This also applies to pitchers who have a full mix but show all their pitches right away, giving batters a look at all their weapons the very first time they see them.

So, how do Rockies starters fare in this study? Let’s first start by showing their 2021 performance per times through the order. We’re going to use the very simple OPS allowed, for clarity, and we’re going to skip over the extremely small sample of PAs facing hitters for a fourth time.

Of course, this is influenced by BABIP (Batting Average on Balls In Play), but BABIP is also a reflection of quality of contact allowed in a way, so let’s roll with it and point out the outliers. You can see how much hitters pummeled Germán last season that third time around, with a jump of 300 OPS points boosted by a jump in home runs (we’ll see that later) and a .339 BABIP. Senza allowed his highest OPS the second time through, mainly on the back of an insane .358 BABIP against, and Gomber actually did better the second time around. One small detail with Gomber: that OPS around .800 he allowed the third time through? That was despite a .200 BABIP. Yikes. Freeland is the most typical of the four, with a steady, almost symmetrical decline.

We’ve seen performance, but now let’s see symptoms. Here’s the K%...

the BB%...

and the HR%...

A few details to look at here. The first thing that stands out to me is the massive drop in K% Freeland experiences as he goes through a lineup multiple times (he goes from 25.6%, to 18.7% and finally to a miserable 14.3% the third time around). The second is the big jump in homers Germán and Gomber had last season, with their home run rates basically tripling. I find it interesting that Gomber had his best BB% the second time around, and it checks out with what I recall, as he often had issues getting settled in.

Now let’s check out each pitcher individually, shall we? I’m using Pitch f/X for this, and the links to each player’s page will be there on their names.

Germán Márquez

Aside from the spike in homers, Germán fared basically how you’d expect. His K% stayed mostly stable and his BB% increased a bit each time around. The walks are, in my opinion, a result of the Rockies’ reliance on swings and misses on breaking balls out of the zone. As hitters see him more often, that breaking ball becomes easier to take. Not easier to make contact with, however, because his breaking ball gets extremely high whiff rates regardless. And I say breaking ball, by the way, because while Germán technically has a curveball and a slider, the two pitches are so incredibly similar in terms of velocity, shape and movement that they can almost be counted as one, as I’ve written about before. This is his pitch usage, for those curious:

His velocity also remains basically unchanged, as he averages 95.0 MPH the first time around and 95.2 and 95.1 the second and third time respectively. I’d suggest more curveballs, because it’s his best pitch, but hey. The lack of a reliable changeup with a different movement profile (along with his lack of fastball usage in two-strike counts) is what’s keeping Germán from climbing to the top tier of aces, at the end of the day. I’m praying that cambio finally appears next season.

Antonio Senzatela

First off, let’s admire the fact that Senza walked just 2.4% of hitters he faced a first time, because it’s a hilariously low number. Of the four, Senzatela is the one who has the biggest velocity drop, going from 95.0 to 94.7 and 94.4 that third time around. His pitch usage is interesting because, like Germán, Senza is basically a two-pitch guy (fastball-slider in his case), but the curveball makes an appearance deeper into ballgames:

It’s most pronounced that first time through, but he also needs a different pitch to complement those two good ones he already has. Of note: batters get better at elevating his slider through the game (56.9% GB first time around, 47.3% third time), but also get worse at elevating his fastball for some reason (52.4% GB to 60.5% GB). The swing and miss against him remains basically identical too. I need more changeups or curveballs, please!

Kyle Freeland

Freeland has a basic plan of attack: throw the kitchen sink at ‘em right from the start. His pitch usage is very steady throughout, and he actually gains some velo (91.3 - 91.5 - 91.7). About that pitch usage, though:

Now, a disclaimer: Pitch f/X registers Freeland’s slider as a cutter and his curveball as a slider. Keep that in mind so you don’t get confused. Anyway, aside from throwing more curveballs the deeper he gets into a game, his pitch mix remains very similar, which is probably why his Whiff% drops significantly on every pitch but (you guessed it) his curveball:

The loss of swing and miss in his four-seamer/slider combo is particularly noticeable. By the third time around, batters whiff on just 6.4% of the swings the take at the heater, and the swing and miss on sliders was almost cut in half, from 27.4% to 15%. I’m not sure there’s a solution for this, too, because this approach is what’s helped him bounce back from the horrors of 2019. Maybe a bit less slider usage early in games? I don’t know.

Finally, just a little nugget I wanted to throw in there: Freeland’s changeup goes from getting grounders over 40% of the time the first two times through to just 7.7% the third. Batters slugged 0.923 against it that last time. Yikes.

Austin Gomber

Gomber is similar to Freeland when it comes to low fastball usage, but he increases his usage of secondaries more steadily than the Denver native:

His velocity remains virtually unchanged (91.8 - 91.9 - 91.7), but I’d like to point out something that stuck out at me: his changeup “ages” better than his slider. Batters whiffed less at Gomber’s slider the more times they saw it (40.6% - 32.2% - 31.1%), but the same couldn’t be said for his change, which also generated way more grounders (42.4% - 54.2% - 60%), so that’s neat. As a sidenote, I’m expecting Gomber to throw a lot more fastballs low in the zone next season than this past one. He has the overhand arm slot you’d want for throwing low and his low-spin four-seamer got tattooed in 2021. That’s probably a story for another day, however.

★ ★ ★

2022 ZiPS Projections: Colorado Rockies | FanGraphs

Mainly putting this here because a) it’s difficult to find new stuff with... nothing going on, and b) for the player comparisons. Germán Márquez gets Bartolo Colón as his main comp for his age-27 season, which is really good! Colón went 15-8 with a 3.88 ERA over 188.0 IP for Cleveland in 2000, a season worth basically 5 WAR. Not too shabby.

The Impact of Modern Pitching Staffs on Payrolls | Baseball Prospectus

In short: pitchers are making less and less on average every year, because there are more of ‘em every year, which means the value of individual hurlers has never been lower. Reduce the size of the pitching staff already, MLB.

★ ★ ★

Please keep in mind our Purple Row Community Guidelines when you’re commenting. Thanks!