More Mopping Up
Yesterday began our discussion regarding mop up pitchers, the ones that enter into blowouts and accrue innings to prevent the unnecessary usage of higher-leverage relievers. I asked for some help identifying the mop up men for several teams, as only a few existed off the top of my head, and got great feedback. One of the comments on the thread, from TangoTiger, suggested we apply the formula: GP*(LI/IP). The minimum amount of innings would be set to 40, in order to ensure these pitchers logged ample enough time to qualify.
The LI component, for those unaware, is Leverage Index, developed by the aforementioned TangoTiger. The stat essentially measures the stress level of the situation at hand. An average LI is 1.00, so when dealing with supposed mop up pitchers, of interest are the average LIs for pitchers equal to, or below, the average. Plugging it into the above formula, dividing by innings pitched, and multiplying that quotient by the total number of games pitched should, in theory, help us narrow these mop up guys down. Basically, the lower the number provided by that formula, the more mop-uppy the pitcher.
Below are the results, with the mop up number next to the name. I was only looking for the mop-uppiest pitcher on each team, so certain players with lower scores than others mentioned will not appear below:
Darren O'Day (Angels), 0.251 Seth McClung (Brewers), 0.279 Josh Rupe (Rangers), 0.289 Aquilino Lopez (Tigers), 0.301 Brian Bass (Twins/Orioles), 0.334 Boof Bonser (Twins), 0.343 Gary Majewski (Reds), 0.361 Clay Condrey (Phillies), 0.365 Joel Peralta (Royals), 0.414 Robinson Tejeda (Rangers/Royals), 0.426 Jon Lieber (Cubs), 0.436 Lance Cormier (Orioles), 0.445 Billy Sadler (Giants), 0.464 Ryan Rowland-Smith (Mariners), 0.478 Jason Hammel (Rays), 0.509 Franquelis Osoria (Pirates), 0.514 Luis Vizcaino (Rockies), 0.523 Jeremy Affeldt (Reds), 0.531 Buddy Carlyle (Braves), 0.543 Nick Masset (White Sox/Reds), 0.551 Jesus Colome (Nationals), 0.558 Cory Wade (Dodgers), 0.603 Doug Waechter (Marlins), 0.624 Brian Tallet (Blue Jays), 0.627 Chris Sampson (Astros), 0.660 Mike Timlin (Red Sox), 0.660 Mike Adams (Padres), 0.763 Jensen Lewis (Indians), 0.788 Edwar Ramirez (Yankees), 0.799 Santiago Casilla (Athletics), 0.926 Juan Cruz (Diamondbacks), 0.969
This is the list generated by the suggested formula. If there are players above that you disagree with, please comment below your case for disagreement, as I would like to have a concrete list to evaluate for Monday. Then, we can identify which mop up pitcher was the best at his job in 2008, and get cracking on sending he, or his agent, an official award. Also, remember, not every team may have a specific mop up man, so some of the players on this list, especially towards the bottom may not belong in the discussion. Curious to hear your thoughts.
13
I see 2 Reds but no Mets or Cardinals… I guess there wasn’t a definitive Met or Card for the list?
James, nope! The 2 Reds is misleading because one was Masset, who was on both CHW and Cin. As far as the Mets and Cards, nobody in the top 50. Schoeneweis was the first Met at #105, and Chris Perez for the Cards at 115.
Eric – including Masset I count 3 Reds, haha: Affeldt, Majewski, Masset.
Obviously I have nothing better to do on a Saturday afternoon.
Oh, wow, so yeah the Reds had quite the bit of mopping up to do.
Doug W. certainly didn’t seem to fit this for the Marlins, but I guess I could have missed him being designated to that.
Yeah, keep in mind that these are merely the results of that formula. From looking at the numbers, though, Waechter fit the bill for the Marlins moreso than anyone else. Of Fish relievers with 40+ IP, he was tied with Logan Kensing for the lowest average Leverage Index, at 0.82, but had 63.1 IP to Kensing’s 55.1. It could have been either, but from my eyes, it’s either one or the other. Whomever you feel more comfortable deeming the mop up Marlin man, let me know.
Just more evidence that the Arizona Diamondbacks are criminally underusing Juan Cruz.
Edwar Ramirez was definitely not the mop up guy in 2008. Yanks didn’t really have a true long man this past season since they never broke camp with one and went through so many injuries. But if I had to pick one that was used in that role, it’s definitely Dan Giese. Ramirez was frequently used as a lefty specialist (even though he’s right-handed) and in the 7th inning.
Hoc, the reason Giese didn’t make the cut is that he only pitched 26.2 innings in relief, which, to me, is not nearly enough to qualify as a mop-up guy. Even if we concede that it was he, and not Ramirez, who was the Yankees mop up man, Giese would not qualify for the Swiffer Mop Up Man of the Year Award.
Yeah, no Yankee mop-up guy reached 40 innings in that role last year. They used mostly Ohlendorf and perhaps Hawkins too as the mop-up guy(s) in the first half of the season, but both of those players were traded away by midseason. Therefore, Dan Giese was called up and sort of took over the role after that, with some mop-up help from Chris Britton as well. But Edwar Ramirez was certainly not a mop-up guy. He was used somewhat often in close games anywhere from the 6th through 8th innings. They even tried him for the LOOGY role occasionally even though he isn’t a left-handed pitcher.
I would argue that the mop-up man for the Giants last season was Yabu, not Sadler. Sadler got thrown into a lot of low leverage situations because he wasn’t very good and needed practice. Yabu was the guy to call on when the starter came up short and the other arms needed rest.
For what it’s worth, Bill Slayback’s 0.067 for the 1974 Tigers is the lowest G*(LI/IP) in the database for those with 40+ innings in relief. Interestingly, Slayback’s 3.70 RA/9 was slightly better than the Tigers’ bullpen as a whole (3.73), but apparently back then if you weren’t one of the predesignated ace relievers on certain clubs you weren’t going to see a whole lot of meaningful action, so it’s not surprising then that on a staff where 36% of the relief innings were thrown by one man (John Hiller), no other reliever posted an inLI above 0.45. Things in that year were toughest in Los Angeles when there were only 89 relief innings to divvy up between men not named Charlie Hough or Mike Marshall. I’m going to go out on a limb and say that no two teammates have monopolized relief innings to quite that extent in any year since.
Also of interest is Johan Santana leading the majors in G*(LI/IP) in 2000 by a healthy margin of .129 to Kevin Beirne’s .188.
I personally would argue that this list is totally meaningless. It fails to show whether a “mop-up man” is able to give his team a chance to win. Thus, I propose introducing WPA/LI (I don’t quite know how), and formally introducing the stat to Fangraph’s repertoire.
I was browsing the team page, to see if “mop up” opportunities came more often to the bad teams, and I was surprised to see that some of the worst teams had the highest pLI. It appeared that there really was no correlation between winning percentage, or WPA for that matter, and pLI.
How about the highest G* (LI/IP) to see which reliever had the toughest time? Also to suggest an answer for a previous comment is maybe the bad teams have higher LI because the other team always has someone on base
Having watched a lot of Astros games, I would disagree with classifying Chris Sampson as a mop up pitcher. Did you differentiate Sampson’s innings between relief and starting? Sampson began the season as a starter in the rotation. He didn’t pitch well in that role. Close to mid-season he was shifted to the bullpen, where he performed very well. As his good innings piled up in relief, he became more of a choice for late inning work in close or tied games. Since I ususally think of a mop up pitcher as the pitcher of choice when the game seems out of hand in the 6th inning or earlier, I don’t recall Sampson with that role. The Astros had several pitchers who had that role at times, two of whom were eventually cut (Borkowski and Villareal).
However, if you pitch the same amount of innings as games, the result is always equal to your LI. Just thought I’d throw that out there. Maybe add an exponent on something. Otherwise, very nice!
Villareal (0.418) and Borkowski (0.469) each failed to meet the minimum number of innings required. If you counted them as a tandem, they’d rank about twelfth on this list.
Wouldn’t it be more effective to somehow include ((WPA+)+(WPA-)) as a way to measure total mop-uppy-ness? That way, we capture how much of an impact a player had. And what about instituting a similar stat for hitters? I’m into creating little excels on this stuff, and I want to institute that. Thanks in advance.
http://www.fangraphs.com/forums/showthread.php?p=3935#post3935
I would love help there as well.
I was surprised to see Affeldt’s name on that list. Could that have perhaps been overlooked when Dave was singing his praises here a few weeks ago?
Either way, it would still seem as though the Giants got him at below-market value, so it still looks like a solid move.
Eric, RE:Marlins
I looked through the game logs and for the first few months of the season, it appears Waechter was the most mop-up like. The problem is that he started being used more for 1 or less than 1 inning later in games. BECAUSE Mark Hendrickson was banished from making a start and became the mop up man after he started opening day, oh those Marlins how I love thee so.
Guys, for those discussing WPA and WPA/LI, yes these will be used when judging which ones were the best, but they are not that effective in determining WHICH pitchers WERE mop-up men. To do that, the LI is more important as it measures the cruciality level of the situation. Tomorrow I will narrow down the list and we can discuss which mop up man deserves the Swiffer Presents Fangraphs’ Mop Up Man of the Year Award.
And CJ, yes, Sampson’s relief innings only were taken into account. Giving your team a chance to win, as Sam suggested, defies the entire point of this, as a) we are just attempting to identify these pitchers, and b) mop up men tend to enter when the game is out of realistic reach for either team.
RRS was a starter for a lot of the year – now, I don’t know what his inning breakdown was, but after being (arguably, but I would not say a mopup) reliever, he was converted to being a starter and was used as a starter by the team for most of the rest of the year.
RRS does suggest an interesting question, however – what about the 4th/5th starter types who are referred to (framed by their team) as starters, and who are usually used as staters, but are also brought in for particular mop-up situations when an offday or their durability allows this to happen?
So the Giants just paid $8M for two years of the Reds mop up man. This analysis is making this signing look really stupid.
And the Giants other reliever target is the DBacks mop up man Juan Cruz. Brilliant!
Sabean is in his little happy place.
I going with the yankees all the way.. Go NY!!!