By design, starting pitchers are going deep into games less often. Instead, teams are turning to relievers earlier than ever before.
This philosophy has considerable merit, especially with more clubs building bullpens capable of repeatedly delivering quality innings. Why start the sixth inning with a back-of-the-rotation arm when a better option is warming up in the bullpen?
Mike Petriello of MLB.com agrees. He recently identified six starters who may benefit from short, successful starts, rather than longer, mediocre outings.
Petriello’s candidates perform well early, but fall off the proverbial cliff after facing a lineup twice. Specifically, their opponent’s weighted on-base average (wOBA) rose above league-average when they saw a batting order a third time.
For those not familiar with wOBA, it is similar to on-base percentage (OBP). However, wOBA gives increasing value to extra-base hits rather than viewing all on-base events equally. If you would like to know more about wOBA, FanGraphs and MLB.com each provide deeper explanations, which are excellent.
Petriello did not discuss any members of the Seattle Mariners, so I decided to apply his wOBA approach to Seattle’s current stable of starters. Are there any candidates worth considering for an early hook?
Before going any further, a few notes about the following table.
With the exception of Erasmo Ramirez, the period covered for each starter includes the 2016-17 seasons. I went back to 2015 for Ramirez because he made just one start for the Tampa Bay Rays in 2016.
Included below are game starts (GS) and the opponents’ wOBA each time a pitched faced the lineup. At the bottom, the MLB average for each time through the order. Highlighted frames were above league-average.
For your reference: league-average wOBA for a reliever during his first time through the order is .309.
|wOBA By Time Thru The Order|
Since Felix Hernandez and Ariel Miranda have struggled out of the gate during the last two seasons, I am not going to discuss the duo. The same applies to James Paxton, who has been strong throughout games.
While seeing Paxton flourish is a reassuring sign for Mariner fans, the fact both Hernandez and Miranda have been unreliable early casts a shadow of doubt over the rotation going into next season. Compounding matters, Moore and Gonzales are unknown quantities, but expected to contribute in 2018.
That leaves us with two candidates to review — Mike Leake and Ramirez. Both were in-season acquisitions by general manager Jerry Dipoto last year with Ramirez returning to the club that originally signed him as an amateur.
Leake is an interesting case. Although viewed as an “innings eater,” his late-inning performance since 2016 suggests otherwise. The Mariners could potentially gain an advantage by replacing the 30-year-old before he craters in games.
Some Mariner fans may be surprised to see how well Ramirez fared when facing lineups three times. Perhaps, this is one of the reasons Dipoto orchestrated a Seattle reunion with the 27-year-old.
Okay, reviewing a starter’s history certainly helps shed light on when to make a call to the bullpen. But when exactly in games have Leake and Ramirez fallen off?
The following table helps answer that question by illustrating each pitcher’s earned run average (ERA) through the fifth inning and subsequent frames.
|ERA by Innings Pitched|
|Pitcher||Overall||Thru 5 IP
Please note, I did not include Ramirez’s ERA for the seventh inning because he faced just 28 batters. My cutoff for both he and Leake was at least 50 hitters in an inning.
While Ramirez performs well through the order three times, his success evaporates after the fifth inning. This suggests pitch inefficiency leads him to see lineups three times by the fifth frame.
Leake has proven he can go longer than Ramirez. But the seventh inning has been the right-hander’s downfall in recent seasons.
Does this mean manager Scott Servais should automatically pull these two veterans strictly based on historical data?
No, when the skipper is mulling a decision, he will likely rely on a blend of thoughtful data-based analysis and discussions with pitching coach Mel Stottlemyre prior to and during game action.
Still, the statistical history of both Leake and Ramirez should not be overlooked. Making a change before they run aground would prove beneficial for the Mariners this upcoming season.
As with defensive shifts, the data we have discussed is not predictive, but the numbers do suggest probability. Just as teams position defenders based on the likelihood of where a ball will be hit, preemptively replacing a starter could prevent a game changing meltdown.
Sure, there will be times when Leake or Ramirez are sailing along and should remain in a game longer than the stats say. That is why there is a manager to make the hard calls and where the eyeball test and talking to the pitcher come into play.
On the other hand, the Mariners appear destined to teeter between fringe contender and also-ran in 2018. If that is the case, the club can ill-afford sticking with at-risk starters too long and missing opportunities to gain a competitive edge in games.