August was a whole new ballgame for relief pitching.
First, Torey Lovullo talked about a problem:
“We’ve got the find a way to get those outs late in the game, leverage or no leverage, it doesn’t matter. When you come out of the bullpen you need to get outs. And if you’re not we’ve got to make some sort of adjustment.” — Torey Lovullo 5 August, earlier in the day.
How I heard/interpreted his comment was that when a closing pitcher enters a game (in a save situation OR a non-save situation), he needs to pitch well. Pitching well in non-save situations was an issue with Mark Melancon. This season, in 13 games he entered the game with the score tied. At least one run was scored against him in 11 of those games.
Later that day, Torey Lovullo announced a different way to decide who closes games.
“We’re gonna back into a situation where we don’t have one closer at this point. Had a discussion with Mark about making this move and he understood. He didn’t necessarily like it but he understood for the good of the team. If it was a decision I was making then he was going to support it. For right now, it’ll be a fluid situation who it will be. It will be undetermined. It will mostly be matchup based.” — Torey Lovullo, 5 August later in the day.
How I heard/interpreted his comment was August would be different. Instead of a primary closer, the closing pitcher would be determined by matchups against with that day’s batters. Two thoughts follow:
- I wondered whether the advantage of better matchups would exceed the specialized skill of a primary closer.
- Who Torey picks for each game’s closer would impact all relief pitching because in planning a game, Torey Lovullo said he works backwards from the ninth inning.
Did the Diamondbacks transform a closer problem into a competitive advantage by closer-matchups? The answer is in the data for relief pitching in August. August was a new ballgame!
Who was the best pitcher in the last inning?
The following table shows results for the last Diamondback inning pitched in each game during August. The table shows Mark Melancon, Ian Kennedy, and a group of 6 other relievers (I excluded last inning relief pitched by catcher Carson Kelly).
Saves. Ian Kennedy had more saves and more save opportunities than Mark Melancon (5 of 6 vs 4 of 5). During August, the others had no save opportunities, although for the season they collectively had zero saves in 6 save opportunities.
On-Base-Percent (OBP)-against. Ian Kennedy had a better OBP-against than Mark Melancon (.278 vs .367). As a strong indication that matchups matter, the other 6 pitchers had the best OBP-against (.214).
Got-The-Job-Done (GTJD). Relievers GTJD if they allowed zero earned runs and allowed zero inherited runners to score. Ian Kennedy had a better GTJD than Mark Melancon (78% vs 75%). The other 6 pitchers has a GTJD of 75%.
Allowed Built Innings (when 5 batters reach the plate). Ian Kennedy had a lower (which is better) percent of allowed built innings (33.3% vs 37.5%). As a strong indication that matchups matter, the other 6 pitchers had 27.3% allowed built innings.
How good was relief pitching in the ninth inning?
As a team in August, the Diamondbacks .319 wOBA was slightly worse than the .310 average in the Majors per Baseball Savant.
In the ninth inning of August games, two Diamondback pitchers had wOBA better than August’s .310 average in the Majors (3 pitchers with 3 PAs each were not mentioned due to small sample size):
- Mark Melancon, .267 wOBA in 25 PAs.
- Ian Kennedy, .303 in 40 wOBA PAs.
Three thoughts/observations follow:
- In the context of his overall .344 wOBA in August, Melancon’s .267 wOBA in the ninth was encouraging because the Diamondbacks have a contract with Melancon for next season.
- Most pitches for Kennedy (96%) and Melancon (69%) were thrown in the ninth inning.
- Next season, Ian Kennedy will be a free agent.
How good was relief pitching in the eighth inning?
As a team in August, the Diamondbacks .340 wOBA was significantly worse than the .298 average in the Majors per Baseball Savant.
In the eighth inning of August games, two Diamondback pitchers had wOBA better than August’ .298 average in the Majors(4 pitchers with 2 or 3 PAs each were not mentioned due to small sample size):
- Edwin Uceta, .126 wOBA in 7 PAs.
- Joe Mantiply, .154 in 30 PAs.
Two thoughts/observations follow:
- This season Joe Mantiply was named as the Diamondbacks’ All-Star.
- Joe Mantiply pitches left handed, and Edwin Uceta pitches right handed.
How did relief pitching in August compare to previous pitching?
As a team the Diamondbacks relief pitching (position = RP per Baseball Savant) wOBA was worse in August compared to previous pitching (.334 wOBA in August vs .318 wOBA in the season through July). Although there were many factors that impacted that comparison, especially different relievers pitched, it’s very possible that at least part of the difference reflected the change in game planning caused by the new approach for picking the closer. In other words, choosing the closer based on matchups possibly made overall relief pitching wOBA worse, but other factors made it unclear.
The following table shows the individual pitcher’s wOBA in August compared to previous pitching. Moronta’s previous pitching was for the Dodgers and the August pitching was Diamondbacks only (excluded August games for Dodgers).
The first six pitchers in the table had an wOBA better than .310, August’s average in the Majors. It is significant that all six pitchers except Ian Kennedy improved in August compared to earlier in the season. My conclusion is that choosing the closer based on matchups likely made the best six relief pitchers’ wOBA better.
Torey Lovullo needed his closer to get outs in all late game situations. The problem was when his closer entered tied games, he allowed runs in 11 of 13 games. Therefore in August he implemented closer by committee. One possible benefit was better matchups that would improve relief pitching.
In August, as a team, Diamondbacks pitching in the ninth inning was nearly average in the Majors, while Diamondbacks pitching in the eighth was significantly worse than average in the Majors.
In August, who was the best pitcher depended on the situation:
- In the last inning pitched in each game, Ian Kennedy was the best pitcher.
- In the ninth inning, Mark Melancon was the best pitcher.
- In the eight inning Joe Mantiply was the best pitcher.
In August, the impact of closing by committee was likely positive:
- When Ian Kennedy’s pitched in the last inning of a game, the results appeared better than the alternative.
- The six pitchers with the best wOBA improved compared to earlier in the season.
- Overall relief pitching wOBA worsened compared to earlier in the season, but that worsening may have been due to other factors.