The idea seemed logical.
With an eye toward the future, the National League wild card game in particular, the Pirates paired Gerrit Cole with Francisco Cervelli for an Aug. 12 tilt in St. Louis.
In most circumstances, a battery consisting of a team’s ace-in-training and its No. 1 catcher would be the obvious choice. Not so for the Pirates and manager Clint Hurdle, who had matched Cole with backup Chris Stewart in 19 of his first 22 starts of 2015.
In fact, after Cole and Cervelli played pitch and catch three times in the opening month of the season, Stewart had handled the former No. 1 overall draft pick 18 straight times — until that aforementioned game at the Cardinals. Prior to that Wednesday evening, Cole had thrown more to Tony Sanchez (four games) than Cervelli (three) in his three-year Pirates career.
Cole hasn’t exactly been terrible since that switch. Counting his start last Thursday at Miami, when Cervelli was ejected in the third inning, Cole has a 3.86 earned-run average, allowing 15 runs and 32 hits in 30 1/3 innings since Aug. 12. During that time, opponents have hit .276, with a .315 on-base percentage and a .379 slugging percentage.
That’s OK for a middle-of-the-rotation starter, but not for a man with the talent of Cole.
Compare those results to his first 22 starts of the year: 2.39 ERA in 143 1/3 innings, opponents hitting a feeble .240/.292/.333. Those results had Cole on track to be a Cy Young Award finalist, but now that’s looking more uncertain.
This step back can’t be entirely pinned on the catcher swap. Maybe his real-life results are simply regressing toward his Fielding Independent Pitching, or FIP, which at 2.70 still remains slightly higher than his 2.64 ERA.
Or maybe he’s tiring in the midst of what will be his most strenuous season. Cole’s career high in innings pitched is 185 1/3 in 2013, although 68 of those were accumulated in Triple-A. This year, he’s at 173 2/3 MLB frames, with Hurdle conceding last week that “(h)e’s going to get to a place he’s never been before regardless.”
With the Pirates clinging to hope that they can make the NL Central race interesting into the final days of the season, there’s little chance that Cole skips a start the rest of the way. But if a few days off isn’t feasible, what about going back to Stewart, who has caught more of his pro starts (30) than anyone?
As is often said in the world of statistics, correlation does not equal causation, but I have to admit the success of the Cole-Stewart connection makes me think there’s something to it. To increase the data set, I looked at the three catchers to work the most Cole starts over the past three seasons:
(A quick primer on the tOPS+ metric listed at the right side of the graphic: It measures Cole’s performance relative to his entire career, with 100 being exactly average. Cole’s 89 tOPS+ when Stewart catches indicates that opponent OPS is 11 percent lower than his career norm in those games.)
I fully expect that Cole’s performance when teaming with Cervelli will improve with a larger sample size, but maybe there really is a reason that the Pirates’ top arm does demonstrably better when Stewart sets up behind the plate.
The concept that a catcher can have an impact on pitching success, presumably via pitch-calling skill, is probably as old as the sport itself. However, as Keith Woolner and several others have shown over the years, so-called ‘catcher ERA’ is highly variable from year to year, and thus is probably more random than meaningful.
But those studies honed in on a catcher’s purported impact on an entire staff, not an individual pitcher. When it comes to one person working with another, I would expect there’s room for a unique synergy that wouldn’t exist in any other pitcher-catcher combination.
Chris Stewart has caught 30 of Gerrit Cole’s 68 career Pirates starts. — GETTY
In looking for differences over the past month or so, there is some variation in pitch usage. According to BrooksBaseball.net, Cole has been throwing his sinker and curveball about five percent more each since Cervelli took over, while four-seam fastball usage has dropped from a little more than half of his pitches to about 42 percent.
That difference might not all be on the catcher, as scouting reports and pitch effectiveness on a given night can affect selection. Another interesting find: Cole’s horizontal release point has gotten closer to his body over the course of the season, so maybe that has changed the character of some of his pitches — and thus the way his catchers are calling the game.
Overall, though, it’s difficult to tell what’s real and what’s not after a handful of starts.
I would be more comfortable judging the decision to go with Cervelli on offensive production than whatever nebulous factor I’m trying to tease out of the PITCHf/x data.
Entering Wednesday’s game in Milwaukee, Cervelli was hitting .301/.376/.420, very good numbers for the position and about 20 percent better than the league-average OPS. Stewart, on the other hand, has posted .280/.311/.333, or approximately 40 percent worse than Cervelli.
Yes, Stewart has put the bat on the ball, but he’s much less likely than his fellow catcher to drive the ball in the gap or draw a walk. For me, that would be enough to write Cervelli’s name on the lineup card for any variety of must-win game in October — unless Cole had an objection, that is.
In any case, Sunday night in St. Louis will tell us a lot about how the Pirates are feeling about who’s catching Cole.
September 3, 2015
Matt’s Stats: Gerrit Cole, with or without Stew
59by Matt Gajtka • Features • Matt's Stats, Pirates
The idea seemed logical.
With an eye toward the future, the National League wild card game in particular, the Pirates paired Gerrit Cole with Francisco Cervelli for an Aug. 12 tilt in St. Louis.
In most circumstances, a battery consisting of a team’s ace-in-training and its No. 1 catcher would be the obvious choice. Not so for the Pirates and manager Clint Hurdle, who had matched Cole with backup Chris Stewart in 19 of his first 22 starts of 2015.
In fact, after Cole and Cervelli played pitch and catch three times in the opening month of the season, Stewart had handled the former No. 1 overall draft pick 18 straight times — until that aforementioned game at the Cardinals. Prior to that Wednesday evening, Cole had thrown more to Tony Sanchez (four games) than Cervelli (three) in his three-year Pirates career.
Cole hasn’t exactly been terrible since that switch. Counting his start last Thursday at Miami, when Cervelli was ejected in the third inning, Cole has a 3.86 earned-run average, allowing 15 runs and 32 hits in 30 1/3 innings since Aug. 12. During that time, opponents have hit .276, with a .315 on-base percentage and a .379 slugging percentage.
That’s OK for a middle-of-the-rotation starter, but not for a man with the talent of Cole.
Compare those results to his first 22 starts of the year: 2.39 ERA in 143 1/3 innings, opponents hitting a feeble .240/.292/.333. Those results had Cole on track to be a Cy Young Award finalist, but now that’s looking more uncertain.
This step back can’t be entirely pinned on the catcher swap. Maybe his real-life results are simply regressing toward his Fielding Independent Pitching, or FIP, which at 2.70 still remains slightly higher than his 2.64 ERA.
Or maybe he’s tiring in the midst of what will be his most strenuous season. Cole’s career high in innings pitched is 185 1/3 in 2013, although 68 of those were accumulated in Triple-A. This year, he’s at 173 2/3 MLB frames, with Hurdle conceding last week that “(h)e’s going to get to a place he’s never been before regardless.”
With the Pirates clinging to hope that they can make the NL Central race interesting into the final days of the season, there’s little chance that Cole skips a start the rest of the way. But if a few days off isn’t feasible, what about going back to Stewart, who has caught more of his pro starts (30) than anyone?
As is often said in the world of statistics, correlation does not equal causation, but I have to admit the success of the Cole-Stewart connection makes me think there’s something to it. To increase the data set, I looked at the three catchers to work the most Cole starts over the past three seasons:
(A quick primer on the tOPS+ metric listed at the right side of the graphic: It measures Cole’s performance relative to his entire career, with 100 being exactly average. Cole’s 89 tOPS+ when Stewart catches indicates that opponent OPS is 11 percent lower than his career norm in those games.)
I fully expect that Cole’s performance when teaming with Cervelli will improve with a larger sample size, but maybe there really is a reason that the Pirates’ top arm does demonstrably better when Stewart sets up behind the plate.
The concept that a catcher can have an impact on pitching success, presumably via pitch-calling skill, is probably as old as the sport itself. However, as Keith Woolner and several others have shown over the years, so-called ‘catcher ERA’ is highly variable from year to year, and thus is probably more random than meaningful.
But those studies honed in on a catcher’s purported impact on an entire staff, not an individual pitcher. When it comes to one person working with another, I would expect there’s room for a unique synergy that wouldn’t exist in any other pitcher-catcher combination.
Chris Stewart has caught 30 of Gerrit Cole’s 68 career Pirates starts. — GETTY
In looking for differences over the past month or so, there is some variation in pitch usage. According to BrooksBaseball.net, Cole has been throwing his sinker and curveball about five percent more each since Cervelli took over, while four-seam fastball usage has dropped from a little more than half of his pitches to about 42 percent.
That difference might not all be on the catcher, as scouting reports and pitch effectiveness on a given night can affect selection. Another interesting find: Cole’s horizontal release point has gotten closer to his body over the course of the season, so maybe that has changed the character of some of his pitches — and thus the way his catchers are calling the game.
Overall, though, it’s difficult to tell what’s real and what’s not after a handful of starts.
I would be more comfortable judging the decision to go with Cervelli on offensive production than whatever nebulous factor I’m trying to tease out of the PITCHf/x data.
Entering Wednesday’s game in Milwaukee, Cervelli was hitting .301/.376/.420, very good numbers for the position and about 20 percent better than the league-average OPS. Stewart, on the other hand, has posted .280/.311/.333, or approximately 40 percent worse than Cervelli.
Yes, Stewart has put the bat on the ball, but he’s much less likely than his fellow catcher to drive the ball in the gap or draw a walk. For me, that would be enough to write Cervelli’s name on the lineup card for any variety of must-win game in October — unless Cole had an objection, that is.
In any case, Sunday night in St. Louis will tell us a lot about how the Pirates are feeling about who’s catching Cole.