Starting Pitching Is Important, Too

It seems a bit silly to write a piece extolling the virtues of starting pitching in the playoffs. Everyone knows starting pitching is important. Those guys pitch three times as many innings during the regular season as their relief counterparts. They win almost all the Cy Young awards — and sometimes MVPs, as well. The narrative this postseason, however, has seen the importance of starters take a back seat to a collection of guys who — either because they’ve lacked the ability to pitch every five days or, otherwise, proved unable to turn over a lineup — have been subsequently moved to the bullpen. Relievers are great — and they’re obviously exerting a tremendous influence on the current postseason — but the starting pitcher’s impact on a game is still great, no matter how early the bullpen comes in to save the day and earn our praise.

Earlier this week, Dave Cameron discussed how bullpen usage was killing offenses this postseason. Teams were shaving off more than an out per game from the starters, who were averaging a little over five innings per start. Going to good relievers earlier and avoiding pitching fatigued starters without having to use the typical below-average long man has kept offense down. On the other hand, five innings of work still represents more than half the outs in a baseball game and it’s the starters who are recording those outs.

During the regular season, the average game score for a start was 51. For those unfamiliar with game score, a pitcher starts at 50, earning a point for every out — two if it’s a strikeout — and losing one for every hit and every walk. A pitcher is docked four points for every earned run (and two for unearned runs). Pitchers also get a bonus for getting into the latter stages of a game, getting two points for every inning completed after the fourth. By leaving a game earlier, a pitcher opportunity for recording an impressive game score is neutralized — although as Jeff Sullivan noted, pitchers are simply a lot better in the playoffs than the regular season.

This year’s postseason starters averaged a game score of 57.5 during the regular season. In the postseason, that same group’s the average game score is 53.0 — which decline basically marks the difference between coming out after 5.1, as opposed to 6.0, innings. As far as results go, starters are performing roughly the same despite facing better hitters. All of this is to say that good starters are generally doing good-starter things while they’re on the mound.

We’ve heard a lot about how good bullpens and good bullpen management are changing the game in October. That seems true, but it’s had little to no effect on later innings. The graph below shows how often teams won when leading heading into an inning during the regular season and during the postseason this year.

We’ve only played 26 games (before NLCS Game 5) this postseason. Nevertheless, those games (on average) have played out almost exactly as they did during the regular season, with teams blowing leads at the same rate. Where we see the difference is in the middle innings. We can chalk some of this up to aggressive bullpen use, but 46 of 52 starters pitched into the fourth inning, while 60% of starters this postseason have completed five innings. Is some of this noise given the sample? Sure, maybe all of it is, but when starters have performed well, their teams have won, and when one starter has outperformed the other, the better starter’s team has won. The bullpen can come in and take over the game, but the starter needs to provide a lead first. For winning teams, that is exactly what has happened.

The graph below groups game scores into three buckets: bad (40 and under), decent (41-59), and good (60-plus). The winning percentages below illustrate the importance of getting a good performance from a team’s starting pitcher.

Note: After last night’s game, the 60-plus group is now 13-5 and the 41-59 group is 11-12.

So what happened in those cases when a starter recorded a good game score and still lost? In two cases, it was merely a product of a pitchers’ duel: Madison Bumgarner just got by Noah Syndergaard in the NL Wild Card game, while Johnny Cueto and Jon Lester matched zeroes in Game 1 of the NLDS before Javier Baez’s solo shot proved the difference in a 1-0 Cubs victory. Poor Matt Moore pitched eight strong innings in Game 4 of the NLDS before the Giants bullpen gave up four runs in the ninth to send the team home. In Game 1 of the ALCS Marco Estrada pitched eight strong innings, but went down to a strong performance by Corey Kluber and the Cleveland bullpen. Max Scherzer, meanwhile, pitched well, but not quite well enough, in Game 5 of the NLDS.

When starters have pitched well, their teams have generally won. Conversely, when starters have pitched poorly, their teams have lost. Only three times this postseason has a pitcher recorded a bad start only to see his team win. One game has already been discussed: John Lackey was the starter who opposed Matt Moore before the Cubs offense came alive in the ninth. Aaron Sanchez, meanwhile, gave up six runs in Game 3 against the Rangers, but the Blue Jays offense scored six runs in the first nine innings and tacked one on in the tenth to advance to the ALCS. In Game 2 of the NLDS, Tanner Roark lasted just 4.1 innings, but Rich Hill did the same — conceding more runs in the process — en route to a Nationals win.

The numbers above don’t even consider how well the other starting pitcher started. When we compare the starters to their opposition on the mound, we see an even clearer picture. Out of 26 games, six had starting-pitcher performances that were relatively equal (i.e. the starters recorded game scores within five points of each other). In those games, the team that had the starting pitcher with the higher score went 3-3. In the 21 other postseason games through last night, one of the pitchers had a game score at least six points higher than the opposing starter. The team that had the pitcher with the higher score has gone 17-4 this postseason.

It’s hard for me to say whether the information above is mind-numbingly obvious or a helpful reminder among the reliever hoopla that starting pitcher performance does actually matter. In the late innings, bullpens are doing exactly what they have been doing all season, shutting down opponents. There have been occasions where aggressive use of the bullpen early in games has helped forge a win, but what has always been true — get the better performance by your starter and win — has still been true this postseason.