by 
The Reds were victims of walk-off wins three times this past week, missing out on a chance to sweep the Marlins and wasting great starts by Sonny Gray and Tyler Mahle against the Cardinals. The Reds bullpen is tied for third in MLB with 30 losses, 11 of which have come at the hands of Raisel Iglesias, a new Reds reliever record. While the offense is still mostly at fault for this disappointing season, the bullpen has been shaky for the past two months.
Blown saves or extra innings losses are brutal ways to lose. They can also lead to subjective evaluations of a reliever (or entire bullpen). If a reliever gives up two runs while his team has a three-run lead, fans will likely be upset initially but will soon forget the individual performance if the team still wins. If the same reliever allows a run in the bottom of the ninth in a tied game, then that performance is looked upon in a much more negative light given the result.
Diving into the numbers, the Reds bullpen has been above average in most metrics. It gets worse when looking at only high leverage situations.
Right off the bat, the bullpen’s ERA nearly doubles in high leverage situations, with FIP going up as well, just not as much. Both stats make for the eighth worst in baseball. One positive is that the K-BB% improves, but when they don’t get a strikeout they have been running into a big issue.
It doesn’t take much to see what the main culprit is; home runs. The 2.2 HR/9 is third worst in baseball, as is the 37.6% ground ball rate. Even though Reds relievers are not allowing much more hard contact, the problem has been the hitters getting the ball into the air. With the current state of the game and the trade-offs hitters are making to hit for more power, it seems that Reds pitchers are falling victim to this in the worst times of the game. So which players specifically have been getting roughed up the most?
None of the specific player performances will tell us if certain people can perform better in higher leverage spots. Studies have been done that show there is no predictive value based on how “clutch” someone has been in the past. If someone is a good pitcher, they should be expected to be good in all types of scenarios. Baseball is a weird sport though, so these numbers help us look at what actually happened and if that is different than what we were expecting.
A lot to unpack here but the overarching message is that the ugly numbers are truly a team effort. The best ERA of the whole group is 4.91, so really everyone has been pretty bad at preventing runs in key situations. Above average home run rates for Iglesias, Lorenzen, Hughes, Stephenson and Peralta have not helped.
Let’s look at the three most used relievers and take a closer look at what has been going on.
Raisel Iglesias
Starting with Raisel, at first glance he appears to be very similar to what we have seen from him the past two years. His FIP is slightly worse (4.39 vs 4.16) but his K-BB%, HR/9 and Hard% are all very close. His ground ball rate is lower than normal, so more line drives and fly balls are not helping. His BABIP is also a bit elevated and could be contributing to the higher ERA.
It is tough to say exactly what is driving the worse than normal results. Unfortunately, this could just be standard volatility that comes with most relief pitchers who are pitching an inning or two at a time. Iglesias probably is not the same pitcher he was a couple years ago. He is also not as bad as his record setting 11-loss season makes him out to be.
Michael Lorenzen
Lorenzen’s pain point is a bit more obvious and falls in line with the overall team issue. Over the past two years, Lorenzen’s HR/9 allowed is 0.9. In his 13 innings of high leverage pitching this year, he is allowing 2.7. While that is a small sample size outlier, it is still a big cause of the ERA and FIP around 6. Another extreme outlier is his 0% LOB rate, which is definitely not sustainable.
On the flip side, Lorenzen has a higher K-BB%, which makes sense since we know he has been effective in some high leverage spots. He also has a super low BABIP, which hasn’t helped him nearly as much as it should. His wOBA allowed of 0.269 is a pretty good showing of Lorenzen’s true abilities, meaning he has pitched well other than allowing too many high leverage home runs (4) which have really hurt his numbers and the Reds.
Amir Garrett
Garrett has only appeared in 11 high leverage innings but has been the most representative of his usual self. His ERA is a little high but everything else is pretty much right where he has been all year. It is tempting to see this and think that Garrett should replace Iglesias as the closer. While that idea may have its merits, it is important to understand that Amir Garrett is not more “clutch” than Iglesias, he has just been a better pitcher this season.
Going forward, Garrett probably should be pitching in more high leverage situations, but we shouldn’t rely on this data to make that conclusion. Even with a rough August, he has been the best Reds reliever over the course of the whole year. With the small number of innings being considered, these numbers are only good for confirming that idea that Garrett has been the best Reds reliever in high leverage situations.
[Featured image: https://twitter.com/Reds/status/1152247780070567940]
Thanks! This was a very good attempt to sort through the Red’s bullpen problems!
What situations are “high leverage”?
The Leverage Index (LI) quantifies each situation based on the affect on win expectancy. Average LI = 1, Low Leverage situations are anything with less than 0.85, and High Leverage is anything with an LI greater than 2.
The full explanation is a bit complicated and can be found here: (https://library.fangraphs.com/misc/li/) if you want to look into it further.
For real game examples. obviously early innings with nobody on base will always be low and bottom of the ninth with the tying runner on 2nd will be high, but I bet that intuition would be fairly accurate when trying to guess any given situation. If you follow the game online at FanGraphs or Baseball Savant, you can check that against the eye test of how important the situation is based on how the win expectancy changes.
The particular challenge for a one-inning reliever is to be difficult for a hitter to handle, in the first (only) time through the order. Mariano Rivera was nigh impossible to hit, when the hitter hadn’t seen him much.
Iglesias by my accounting has yielded runs in 17 of the 58 games he’s pitched in. Brad Hand has yielded runs in 12 of 57 games; Aroldis Chapman, 12 in 55. It doesn’t sound like much, but Iggy is more than 30% worse than the others at the key function of not being scored upon in one inning. Losses can be a bad stat, but not in this case.
Iglesias just isn’t very hard to hit for a hitter seeing him for the first time. Lefties, facing a RH who generally comes from the side or 3/4, have hit 7 HRs in 22.1 innings off him. That is Bad-Bronson level.
We can overthink pitching sometimes; if a guy throws a lot of meatball pitches in the middle of the strike zone, then MLB hitters are gonna hit them, hard. Iglesias just throws too many fat pitches. They need a GL stat, for when you say “Good Lord” upon watching the replay. Iglesias is the King of GL.
How to judge Garrett and David Bell’s bullpen?. Normally, your top pitchers pitch in high leverage situations regardless of handedness. Randy Myers/Aroldis Chapman/ and John Franco didn’t come out against righties. Rob Dibble and Danny Graves and Scott Williamson and Coco Cordero pitched against lefties. Garrett is a glorified Loogie. Bell pulls him against righties. In judging high leverage relievers, I want to know the guys who can get outs, regardless of handedness.