Gerrit Cole And the Pirate Way
Watch Gerrit Cole pitch and you might come away expecting different numbers on the back of his baseball card. He’s 96 mph with the fastball and his third-best pitch looks like this. You’d think he’d be racking up the strikeouts.
But Cole is striking out fewer batters than league-average. And that’s just as he wants it — to an extent. His team, in fact, is probably proud of that statistic. You might even call it the Pirate Way.
That Cole still throws his fastball more than two-thirds of the time and has a 50% ground-ball rate cannot solely be put on the Pirates, of course. Cole says he was just as much of a ground-ball guy in college. And, in an interview late last month, he told me his delivery is “built around” his fastball. “Pretty much my success is built off of fastball command,” he said. “I kind of just let the off-speed pitches show up when they do.”
If he threw the slider, the changeup and the curveball more, Cole would get more strikeouts. That much is known. Those three pitches get three times the whiffs of his fastballs. And they don’t get terrible ground-ball rates, either. Yes, his sinker gets grounders 59% of the time, but his slider and curve both get more than 50% grounders, too. So the restraint — he doesn’t throw any of the three more than 15% of the time — is not just about getting ground balls.
The Pirates’ starters lead baseball in ground-ball rate (by almost five percentage points over the Cardinals), so obviously they stress grounders. But Pittsburgh also is sixth in the league in two-seamer percentage, so the team’s approach might dial down to pitch selection. Simply put, the Pirates like fastballs.
Cole agreed. “That’s all they preach is ground balls,” he said. “Ground balls and fastball efficiency, really, trying to create plane with the fastball. That creates sustainability. When you come up with a faster breaking ball, that’s good. But in the long run, for the guys who pitch for a long time, it boils down to fastball command.”
We’ve seen some of this from other organizations. For example, my colleague Jeff Zimmerman showed that Oakland general manager Billy Beane’s preference for pitchers with command leads to better health. And the idea is simple on one front: You throw the fastball more than any other pitch; make sure you repeat those mechanics so you’re on solid footing for the rest of the game.
But when it comes to efficiency, it’s a question of scale. Here are the top six and bottom six teams in ground-ball rate, and their pitches-per-batter-faced.
| Team | K% | GB% | Pitches/BF |
|---|---|---|---|
| Pirates | 20.1% | 52.9% | 3.754 |
| Cardinals | 20.5% | 48.8% | 3.798 |
| Rockies | 17.1% | 47.3% | 3.783 |
| Dodgers | 21.2% | 46.9% | 3.856 |
| Tigers | 22.6% | 45.9% | 3.943 |
| Dbacks | 19.7% | 45.9% | 3.768 |
| 20.2% | 48.0% | 3.817 |
| Team | K% | GB% | Pitches/BF |
|---|---|---|---|
| Athletics | 18.5% | 38.9% | 3.823 |
| Giants | 20.8% | 41.6% | 3.856 |
| Orioles | 18.3% | 41.8% | 3.872 |
| Royals | 19.9% | 42.0% | 3.898 |
| White Sox | 20.8% | 42.4% | 3.935 |
| Angels | 19.0% | 42.5% | 3.867 |
| 19.6% | 41.5% | 3.875 |
The difference between the top fifth of the league in ground-ball rates and the bottom fifth of the league is .058 pitches per batter faced. (Repeat this with strikeout rate, and the difference between the top tier strikeout teams and the bottom tier is .046 pitches per batter faced.) Of course, most teams have faced more than 5,000 batters at this point, so that’s saved the top ground-ball teams almost 300 pitches as a staff. Divvy that up among various pitchers, and you’re talking about saving starting pitchers between 30 and 40 pitches a year simply by focusing on ground balls.
Of course, the batting average on those ground balls in play is meaningful. After all, what’s 40 pitches saved if you have to face more batters? A strikeout is fascist because doesn’t allow the chance of a ducksnort. The Pirates have the fourth-best BABIP in the league, though, and their aggressive shifting is a big part of that. As Cole said, “For the most part, we call the game and we play the game according to how we are going to position our fielders. It’s been working out for us.”
But there might be a little give in that philosophy. Cole’s 21% strikeout rate in the second half has actually been above-average, and that’s corresponded with a reduction in fastball percentage. If you add his fastballs before July 15th (his first seven starts), you see he threw his fastball 77% of the time. Since then, he’s only thrown one of his fastballs 67% of the time. His slider percentage has gone up from 6% to 21% to eat up most of that difference, as you can see in the chart below. Cole said his slider was his “most consistent” secondary pitch, so this makes sense.
| Before 7/15 | After 7/15 | ||
|---|---|---|---|
| CH | 5% | CH | 2% |
| FA | 3% | FA | 4% |
| SL | 6% | SL | 21% |
| FF | 63% | FF | 40% |
| FT | 11% | FT | 23% |
| CB | 12% | CB | 9% |
By pairing on-field shifting with an emphasis on fastballs — and ground-balls in particular — the Pirates lead the league in fewest pitches per out. The team’s approach is working so far, and it seems well-grounded in the best available research: Ot might keep their pitchers healthier, and at the very least it promotes healthy practices. When it comes to individual pitchers like Cole, though, it’s probably a good idea he’s allowed some give when it comes to the Pirate Way. Because, you know, the batting average on a strikeout is still zero.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
I think there’s something to be said for reigning in a young pitcher during his first year and trying to have him pitched in a controlled environment. I think that in the next year or two, once Cole has his footing in the majors, you’re going to start seeing him throw those strikeout pitches more and dominating some games. This year, the Pirates just need him to control games, but give it a year and I think we’ll have a totally dominant guy on our hands.
Worth noting that Cole’s Strikeout % has been a tick above league average the past 2 months or so, which coincides with his greater use of the four seam and slider, and limited use of the sinker.
I also thought that was worth noting, so I noted it.
Yeah, I completely missed that paragraph when I first read the piece… And the accompanying chart 🙁
Cole has been compared to Verlander since draft day and its funny that people forget that Justin had a 6.00 K/9 in his rookie season with a much higher BB rate and FIP. Cole is a year younger and the focus on not fearing contact has all but solved his previous BB issues. Teach the kid to trust his stuff then take off the leash and let him dominate. As was mentioned, he is already getting more K’s by using his full arsenal.
30 to 40 pitches a year doesn’t sound all that significant, especially if you’re actively sacrificing strikeouts.
I’m wondering:
Imagine a two team, one game league. Team A has a certain pitching philosophy and throws 100 pitches, and team B has a different pitching philosophy and throws 90 pitches. It can only be said after this game (this season) that team B saved 10 pitches with its pitching philosophy COMPARED to team B. This may be true (other factors to consider) only relative to team A. But this is not a necessary truth.
It is possible that team B, using team A’s pitching philosophy could have thrown 120 pitches–a 30 pitch swing from what happened to what could have happened. So there is the fact of what happened and the worst case scenario possibility of what could have happened using the same pitching philosophy as team A.
Team A pitching philosophy = high strikeout/avoid contact; team B pitching philosophy = groundballs. The minimum number of pitches a pitcher can throw in a game is 27 if he throws 1 pitch to each batter to groundout. If a pitcher Ks all batters faced in a game, the minimum number of pitches he will throw is 81. A 54 pitch difference.
But there is also a possible
Ah! I agree on one level. But the Pirates have a league-average strikeout rate. And 300 pitches on a staff level, that’s more meaningful probably. 300 fewer chances to tear something.
It’s interesting to see the difference in pitching philosophies of the Pirates, A’s – and then teams like the Twins. Whereas the former teams seem to be emphasizing ground balls and effective strike throwing, the latter has opted for simply ‘pitch to contact’. Perhaps high control/high groundball pitchers are the latest market deficiency/moneyball strategy, but the Twins philosophy seems to be a huge failure over the pasta few years.
The A’s don’t emphasize groundballs. Their starters have the lowest groundball percentage in the MLB.
Ah OK I misread the statement in the piece regarding Beane’s preference for command pitchers. I guess it makes sense given their stadium and the advantages it gives to fly ball pitchers.
Makes sense for their chasm of a ballpark. Still better than the twins.
Yeah I just hope Twins management doesn’t mess up some of their intriguing talent coming up (Stewart, Meyer, Buerrios, May). Other than Santana/Liriano, I can’t think of a Twins starting pitcher in recent memory to have an above average strikeout rate.
A quibble with your math ,or logic . You say that the top 5 teams in groundballs are averaging .058 less pitches per OUT and then say that they have faced roughly 5000 batters ,and then evidently multiply that .058 by 5000 to get 300 pitches saved .You are treating that .058 number of pitches saved as if it was PER BATTER FACED ,not PER OUT .The correct number of pitches saved would be roughly (118 games X 27 outs = 3186 outs X .058 = roughly 190 pitches instead of 300 ,which makes an already small number quite a bit smaller . So small ,in fact, as to be completely statistically irrelevant when spread out among all of the pitchers on the staff .
“A strikeout is fascist because doesn’t allow the chance of a ducksnort.”
Hahahaha! In the apparent spirit of the Pirates, were you trying to sneak a fast one by the Fangraphs community? Well, it worked for four hours.
Sure there’s a chance – it’s called a wild pitch/passed ball. If the batter were not able to advance on those then your statement would be factually correct. I love the simile, however wrong it is in application.
Eh, the wild pitch is not called a ducksnort. And something like 99.7% of all Ks are outs.
What about the standard deviation on the average pitches per out?
The story you usually hear with groundball pitchers is that they can get quick outs, easy outs, double play balls, etc.. And I think high pitch count innings and games are more correlated to injuries than overall innings or pitches. So a groundball pitcher would be usually close to 15 pitches an inning and 100 pitches per game which seem safe, but a strikeout pitcher would be more likely to deviate and have 30 pitch innings and 120 pitch games.