WAR Lords of the Diamond (Relief Pitchers)
Previous Sessions in the WAR Lords Series:
C (with methodology) I 1B I 2B I 3B I SS I LF I CF I RF
For the next step in my ongoing quest to quantify the greatest Rockies ever by position, I'm revealing the best performers at relief pitcher.
Obviously this is not a comprehensive list of all relief pitchers, but all the pitchers who have made a large contribution to the Rockies as relievers were weighed, measured, and (often) found wanting.
Once again, I'm using Sean Smith's historical WAR database (not Fangraphs) to compile these numbers (with a big assist to the fabulous Baseball Reference; here is the glossary of the terms Smith uses and an explanation of how pitching WAR is calculated.
As I explained in the pitching WAR article (please reference it for more detail on this point), relievers are by and large shortchanged by WAR (which is a counting, not a rate stat). This is because WAR ignores the game's situation and context--a strikeout when the bases are empty in a blowout is counted the same as a ninth-inning strikeout with the bases loaded in a one-run game.
As a result, relievers, whose primary value comes from delivering in high leverage situations, have their contributions diminished by WAR. There are stats out there that measure this well (like Leverage Index and Win Probability Added), but to measure relievers by those stats is both inconsistent with the rest of this series and impossible (the stats don't go back to 1993). In any case, Smith's WAR database remains an excellent way to quantify pitching value, starter or reliever.
Relief pitchers have consistently gained influence in baseball since the 1960s, evolving from the realm of broken down failed starters into a highly specialized field. As Dave Cameron said though:
Relievers are, in general, failed starting pitchers who are given an easier task that their skillset will allow them to handle. They are selectively managed to face hitters whom they have the best chance of getting out, and they get to throw at maximum effort on nearly every pitch, giving them greater velocity over their shorter appearances.
Nearly every starting pitcher in baseball could be a useful relief pitcher. Very few relief pitchers could be useful starting pitchers.
This doesn't necessarily hold true for the rare dominant closer, but by and large this is a good explanation of what a relief pitcher is. Many teams have chosen to bring their stud starting pitching prospects up in the bullpen to get them used to the majors and because it is relatively much easier than starting. Generally back-end relievers have one or two plus pitches that they can throw at maximum effort, often generating excellent results--but they would fail under the rigors of handling a lineup two or three times in a row.
The bullpen is by nature the most fungible piece the team, generating high turnover due to the incredible variability of relief pitching performance. After all, the Rockies only have Huston Street this year that has stayed in the bullpen all year long. While this is a little extreme, the Rockies have had quite a few relievers in their short history.
In order to qualify for consideration for this list the pitchers had to be one of the top five relievers in innings pitched in a given year or throw over 50 innings for the team in a mostly relief role. It's a relatively deep group, with 18 relievers delivering at least 1 WAR for the Rockies.
Relief Pitchers
Note: some of these pitchers were both starters and relievers for the Rockies. I have placed these players in this category subjectively on the basis of where I believe they provided the greatest impact to the Rockies. In case of an average rank tie, career WAR is the tiebreaker. I'm ranking seven players to reflect the fact that the modern bullpen generally uses seven pitchers.
1. Steve Reed
Career WAR: 9.2 (1st)
Top Three Seasons: 5.6, 1995-1997 (1st)
Top Season: 3.1, 1997 (T-2nd)
Average Rank: 1.33
Steve Reed is one of my favorite all-time Rockies. Though he never really closed for the team (15 saves over 7 years with the Rockies), I usually felt pretty secure about a lead that he protected. Reed is a great story, having signed with the Giants as an undrafted 23 year old college free agent, somehow making it to the majors at 27, and carving out a 13 year career out of it.
The Rockies acquired the right-hander with the 60th pick of the expansion draft and he paid immediate dividends for them, taking the ball 64 times in 1993 and solidifying the back end of the Colorado bullpen for five years. His best year was the magical 1995 Wild Card run (one of several excellent bullpen performances that year), in which he was 5-2 with a 2.14 ERA over 84 IP. His K/9 was 8.46 (3.76 K/BB) with a FIP of 3.25 and an incredible ERA+ of 252! All in all it was worth 3.1 WAR--a fantastic reliever season. Street this year has been worth only 1.5 WAR.
Reed left after the 1997 season for free agency, but returned as a 38 year old for the 2003 and 2004 Rockies, where he posted two 1.2 WAR seasons (great for a reliever). Indeed, Reed never posted a negative WAR for the Rockies, a stunning model of consistency in a variable profession. This consistent excellence is what made him the greatest Rockies reliever ever.
Career WAR: 8.8 (2nd)
Top Three Seasons: 5.2, 2005-2007 (2nd)
Top Season: 2.7, 2005 (4th)
Average Rank: 2.67
The Rockies' all-time saves leader with 115, Brian Fuentes was acquired in the Jeff Cirillo trade from Seattle after the 2001 season. The former 25th round draft pick uses a deceptive left-handed delivery and plus stuff to baffle hitters from both sides of the plate, but as many Rockies fans will tell you, it was always an adventure with Fuentes.
Fuentes has had two seasons over 2 WAR of his seven with Colorado, the first of which occurring in 2005, when he posted a 2.91 ERA and a 3.40 FIP over 74.1 IP with a K/9 of 11.02 (K/BB of 2.68) and an ERA+ of 164 on the way to 2.7 WAR. Last year Fuentes had similar success (2 WAR) with a 2.73 ERA (2.24 FIP), 11.78 K/9 (3.73 K/BB), and 168 ERA+ over 62.2 IP.
As all things must, the Fuentes era came to a close (get it?) as he moved out of the Rockies' price range with his fine performance and as a Type A free agent garnered the Rockies what turned out to be Tim Wheeler and Rex Brothers. I think that Rockies fans are happy with the guy we got to replace him, though we certainly thank Brian for his seven years of service.
Career WAR: 5.2 (4th)
Top Three Seasons: 3.2, 1993-1995 (9th)
Top Season: 3.1, 1995 (T-2nd)
Average Rank: 5
Contrary to Steve Reed, I had an irrational hatred of Curtis Leskanic as a young fan, predicting doom and gloom every time that he took the mound. Colorado acquired Leskanic (originally a 1989 8th round pick of the Indians) with the 66th pick in the expansion draft and plugged him right into the bullpen. The right-handed Leskanic generally handled setup duties and generally terrified me when he did so. I was even more mortified when he handled the closer position (20 saves as a Rockie).
The numbers by and large back me up on my fears, though Leskanic's saving grace on this list is his phenomenal 1995 (similar to Reed), in which he also delivered 3.1 WAR. Leskanic had a 6-3 record with 3.40 ERA (2.86 FIP) over 98 IP, 9.83 K/9 (3.24 K/BB), 10 saves, and a 159 ERA+. Most of Leskanic's Colorado years were very average besides his career year in 1995, as he was hampered by his high walk rate and poor FIP.
In any case, my personal torturer was traded by the Rockies after the 1999 season to the Brewers for another reliever, lefty specialist Mike Myers.
4. Bruce Ruffin
Career WAR: 5.7 (3rd)
Top Three Seasons: 4.6, 1993-1995 (3rd)
Top Season: 1.9, 1993 (10th)
Average Rank: 5.33
Bruce Ruffin is yet another original Rockie near the top of this list. The 1985 2nd round draft pick of the Phillies signed with Colorado as a free agent before 1993 and he finished out his career with the Rockies, through 1997. Though he'd primarily been a starter before coming to the Rockies, Ruffin was primarily a bullpen guy for Colorado. He served as Colorado's closer in 1994 and 1996, posting solid seasons both years. In all he finished with 60 saves as a Rockie.
Like Reed and Leskanic, Ruffin was excellent in 1995 as a full-time reliever (1.7 WAR) but his best overall season was 1993 (1.9 WAR), in which he started 12 games (remember, starters are more valuable than relievers!). All in all, Ruffin was very solid for the Rockies over his tenure with the team.
5a. Jerry Dipoto
Career WAR: 4.7 (5th)
Top Three Seasons: 4.4, 1997-1999 (4th)
Top Season: 2.1, 1999 (8th)
Average Rank: 5.67
Jerry Dipoto, now an executive in the Diamondbacks system, pitched for Colorado from 1997-2000. The 1989 3rd round pick of the Indians was traded to Colorado in exchange for Armando Reynoso after the 1996 season. The right-handed Dipoto served as the Rockies' closer in 1997 and 1998, racking up 36 saves as a Rockie.
Dipoto was at his best, though, as a setup man in 1999 (2.1 WAR). Unfortunately, Dipoto's career was derailed by a neck injury before the 2000 season as he competed for the closer's job again, cutting short a promising career.
5b. Jose Jimenez
Career WAR: 4.2 (6th)
Top Three Seasons: 4.2, 2000-2002 (5th)
Top Season: 2.3, 2000 (6th)
Average Rank: 5.67
Jose Jimenez, number two on the Rockies' all-time saves list with 102, was acquired by Colorado after the 1999 season as part of the trade that sent Dave Veres and the late Darryl Kile to St. Louis. Colorado converted the 26 year-old hard-throwing righty from a starter into a reliever. Jimenez made an immediate impact with the Rockies in his new role, saving 24 games with a 3.18 ERA (3.94 FIP) over 70.2 IP with a 182 ERA+ despite only a 5.60 K/9 ratio (suggesting a fluke), earning a WAR of 2.3.
Unsurprisingly, Jimenez never was that dominant again, though he did have two more useful seasons (including 41 saves in 2002) as the Rockies' closer, plus a fourth one that was poor and led to the Shawn Chacon experiment. The Rockies let Jimenez leave after 2003 and within a year he was out of baseball.
Career WAR: 3.8 (7th)
Top Three Seasons: 3.6, 1995-1997 (6th)
Top Season: 2.3, 1995 (6th)
Average Rank: 6.33
Filling out my imaginary bullpen is the Rockies' first closer, Darren Holmes (he did the honors in 1993 and 1995). The 16th round pick of the Dodgers in 1984, Holmes was selected by the Rockies with the fifth pick of the expansion draft and inserted right into the closer's role (he saved 46 games as a Rockie).
Further showing just how special the 1995 bullpen was, Holmes had his career year then (2.3 WAR), going 6-1 and saving 14 games with a 167 ERA+. Once again, the Rockies in 1995 had Reed (3.1 WAR), Leskanic (3.1), Holmes (2.3), and Ruffin (1.7). That's quite a back end right there.
Holmes stayed with the Rockies through 1997, after which he kicked around the league for six more years, never again attaining the same success he enjoyed in Colorado.
Others of note: Mike Munoz, despite having pitched six years for the Rockies, ended up with a net -0.4 WAR (32nd overall) and an injured Roger Bailey on his conscience.
Gabe White owns the best single WAR reliever season (seriously) with his 3.3 WAR 2000. White in 83 IP was 11-2 with a 2.17 ERA (2.71 FIP), 9.00 K/9 (5.60 K/BB), and an ERA+ of 267! He never replicated that success again.
Manny Corpas finished tied with White for eighth with 3.4 WAR, while Taylor Buchholz (2.1) was 14th and Jason Grilli (1.1) was 16th.
Next week I'll tackle starting pitching, and then it will be time to assemble the greatest Rockies team ever.
35 comments
|
1 recs |
Do you like this story?
Comments
Anybody see Fuentes last night
it was like watching a rerun. Get the first two outs then walk, hit guys, hang curves and CUs and lose.
by PinchHitLancePainter on Sep 17, 2009 10:02 AM MDT reply actions
terrible call witht he bases loaded
but yeah i get your point he asked for danger
we seem to have gotten over that HURDLE
"TuLoRocks2008 had the most comments I've seen in any game thread this year (322)...Can such a feat be eclipsed?" ..... Bring it on!!
by TuLoRocks2008 on Sep 17, 2009 3:59 PM MDT up reply actions
Thanks for great post..
I am still reading and learning.!! Yes, I saw Fuentes and am so glad he is the worry of someone else.! Can’t wait to have Street back on Sunday.
Ha...Grilli is 16th despite less than a year with the club
I suppose it’s a good sign that particular front office move is the one I disagree with most.
"I have no special talents. I am only passionately curious." - Albert Einstein
by Andrew T. Fisher on Sep 17, 2009 10:35 AM MDT reply actions
I'd put Fuentes 1st
I liked Reed, too; he pitched really well in a heinous environment for pitchers. But since the overall WAR numbers are so close, I’d move Fuentes up for his time as 8 inning guy and closer. The roles push his # of high leverage innings up a bit over Reed. (For bb-ref users, not I’m not going by aLI, since that appears to at least partially “credit” a pitcher for getting himself into deeper trouble.)
Leave Dexter alone! You're lucky he even performs for you!
BTW, Cameron put up an article on Fuentes in response to his topsy-turvy Angels season
Link.
My first impression is that he seems to have been a lot worse for the Angels than he was here.
Leave Dexter alone! You're lucky he even performs for you!
Interesting
so aLI will calculate a LI based on the depth of the WPA to the 1.000?
Meaning, if he enters the game and it’s a .900 WP% for the Rockies, and the WPA should he get the save would be .100 WPA. Unless he lets runners on, which might drop the lower part of WP% to like .750, meaning that if he gets the save, he gets a .250 WPA for getting the save.
I know leverage and WPA aren’t the same thing, but is that kind of the rationale behind it?
Mike McCoy Status: FREEEEEEEEEEEEEEEEEEEEEEEEE
Seth Smith Status: Bludgeoning Baseballs
Chris Iannetta Status: OH MY GOD GET SOME HITS YOU'RE MAKING ME LOOK LIKE AN ASS
Check out the most recent MLB Transactions on MLB Daily Dish
by Andrew Martin on Sep 17, 2009 2:37 PM MDT up reply actions
Not sure
I’ve never looked that closely at WPA, so I don’t really know how they calculate it. I assume they sum the plays, so in your example, you have to account for the negative WPA of the plays that lower the expected WP to .750.
I believe LI expresses the greater impact of events in a base-out state when the game state changes…that is, a double with bases empty and no outs is worth one value when up by 5 in the 4th inning of a tie game, but something quite different in a tie game in the 9th at home.
aLI sounds like the average LI of each PA in the appearance. So, Morales entered the game in the bottom of the 9th, 0 outs, home team down by 4; the initial LI is 0.5.
The rest of Morales’ evening would go:
2nd PA, LI: 1.0 (runner on 1st, no out)
3rd PA, LI: 2.0 (runners on 1st and 2nd)
For an aLI of 1.17, after an initial LI of 0.5
Betancourt then entered with an initial LI of 2.9 (none out, runners on 1st and 2nd, home team down 3).
Leave Dexter alone! You're lucky he even performs for you!
I made a Counting Rocks about WPA
it’s the Manny Corpas one. The idea is your WPA is the change from when you begin and when you end – so for a pitcher, it’s obvious, but you’ll have several WPA situations for each batter.
Mike McCoy Status: FREEEEEEEEEEEEEEEEEEEEEEEEE
Seth Smith Status: Bludgeoning Baseballs
Chris Iannetta Status: OH MY GOD GET SOME HITS YOU'RE MAKING ME LOOK LIKE AN ASS
Check out the most recent MLB Transactions on MLB Daily Dish
by Andrew Martin on Sep 17, 2009 9:38 PM MDT up reply actions
I take this back for now
Since it seems Smith’s WAR uses leverage index in his pitcher WAR…that should properly credit Fuentes for his role.
Leave Dexter alone! You're lucky he even performs for you!
Yeah, I noticed that Smith factored in leverage into WAR somehow...
I’m just not sure how exactly his calculation works.
Eschew Obfuscation!
Good question
eyeballing Fuentes’s page, I’m not sure.
I would assume you’d take the runs line, add the “DEF” line to adjust for team defensive contributions, then take the difference from the replacement pitcher total. I get 98 runs. Or 103 runs, if the “5” means the defense cost him 5 runs.
He then says the pitcher is “credited with extra leverage halfway between their actual leverage and 1.0” since a closer would not be replaced by a replacement pitcher, but by the next best pitcher. Since Fuentes’ leverage is above 1.0, you’d expect to see some extra credit there. But he has a career WAR of 8.8. To get something above 98 runs to 8.8 wins, your runs to wins multipler would need to be pretty high, over 11 runs per win. I don’t think Coors’ run environment in his time was that high.
So I must be missing something.
Leave Dexter alone! You're lucky he even performs for you!
I attended a Rockies game against the Angels..
the one Ubaldo pitched and the Rockies were down in the 9th. Fuentes came in and the fans around me were talking trash. I turned to the least drunk of them and said, “get ready to be nervous.” Afterwards the usher came up to me and asked if he pitched like that in Colorado.
Order a Purple Row t-shirt and donate to charity at:
Purple Row Cares
Reed measure better than Fuentes in each of my three categories...
none of this is subjective.
Eschew Obfuscation!
That's fine
I don’t think I claimed you were being subjective. I was saying I would make a change before I realized Smith used LI in his WAR calc.
Leave Dexter alone! You're lucky he even performs for you!
I remember Curtis Leskanic
My mom used to call him “Cardiac Curtis”. I believe he loaded the bases every time he come out. I don’t remember him actually blowing games, but I remember him almost blowing them every time. However, I was young, and my memories of that era are generally rosy. For instance, I believe that every time the Rockies were tied or losing entering the 9th at home, Dante Bichette or Larry Walker won or tied the game with a home run. I’d have to look that up though.
by controlled_slide on Sep 17, 2009 10:46 AM MDT reply actions
I remember Leskanic for 2 rumors
For chugging a Mountain Dew or two in the bullpen to get ready, and for supposedly shaving his right arm to cut down on air resistance and improve velocity. He was a character.
Gladly sharing a name with Dexter Fowler!
by ShadowPenguin on Sep 17, 2009 10:49 AM MDT up reply actions
I remember the arm shaving thing too now
I had forgotten about that. I loved that rumor.
by controlled_slide on Sep 17, 2009 10:55 AM MDT up reply actions
plus his intro music was the doors
he changed songs around, but hey – anyone who like Jim Morrison – that has to be worth several pts right there.
and the arm shaving is true…
by RockyMtnCat on Sep 17, 2009 11:05 AM MDT up reply actions
if it's on wikipedia... it must be true:
During his playing days, Leskanic used to keep his right arm shaved as to lower wind resistance.
http://en.wikipedia.org/wiki/Curtis_Leskanic
http://twitter.com/blakebomber
by BlakeBomber on Sep 17, 2009 12:58 PM MDT up reply actions
I remember Holmes
He had one of the filthiest, loopiest hooks ever
Patrick Saunders: I think Kruk is lazy.
My dad still blames Holmes for his inability to stomach the stress of watching baseball.
I’m not sure why, as I was too young to truly comprehend the nuances of the game then….but he complains about Holmes at least five times a summer
"I have no special talents. I am only passionately curious." - Albert Einstein
by Andrew T. Fisher on Sep 17, 2009 12:19 PM MDT up reply actions
My sister still complains
about Shawn Chacon.
Notice he didn’t make the list.
http://twitter.com/blakebomber
That seems far more rational
He was worse, and it was more recent
"I have no special talents. I am only passionately curious." - Albert Einstein
by Andrew T. Fisher on Sep 17, 2009 1:03 PM MDT up reply actions
I wasn't following the team on a daily basis that season
but that must have been pretty brutal. I can’t believe they didn’t give someone else a chance to take over. (I know the answer…)
Leave Dexter alone! You're lucky he even performs for you!
hey jaberwocky, quick question for you
you may have answered this in another WAR lords but why are Sean Smith and Fangraph’s WAR ratings different? Do they calculate them differently or do they have a different definition of what a replacement player is?
They calculate them differently
For position players, the underlying concept is pretty much the same: quantify park-adjusted batting runs. Add in fielding, a positional adjustment, and a value for replacement level, and convert the resulting run total to wins.
For pitchers, there’s one big conceptual difference. Fangraphs uses FIP to set a pitchers runs allowed, whereas Smith uses actual runs allowed, adjusted by the total zone for the teams they pitched for. The total zone values are weighted by the pitcher’s batted ball distribution. Since FIP abstracts from both defense and the order in which events occur (think solo HR’s vs. HR’s allowed with runners on), this approach might make a difference for pitchers who are unusually good/bad with runners on.
I now notice Smith also uses a leverage index for pitchers, so this should affect WAR for closers (and probably nullifies my comment above about Reed vs. Fuentes).
Differences for position players: They use slightly different systems to estimate runs. Smith gears his so the sum of players’ runs won’t exceed their team’s total runs. Not sure where they get their park factors from; I assume Smith calculates his own. Smith adds in baserunning (that is, more than SB/CS, taking extra bases, GIDP’s). You see this really helps a player like Ichiro, who is an excellent percentage basestealer, takes extra bases, and grounds into very few double plays.
For fielding, fangraphs uses bUZR: UZR using BIS data. Smith uses Totalzone, a similar system, but with a different dataset. play-by-play data comes from retrosheet, and retrosheet’s data varies as the scorekeeping on which they rely has changed over time. For seasons before retrosheet had pbp data, total zone estimates batted ball distribution.
For the positional adjustment, I believe fangraphs uses values calculated by Tom Tango; not sure of the years of data that went into this. Since Smith’s WAR covers many more seasons, he does the positional adjustment by decade.
Replacement level for both is generally 20 runs/650 PA, but Smith says he also varies this depending on league strength. I don’t know for which years he does this, or by how much it varies.
Also, converting runs to wins is usually 10 runs/win, but Smith varies it depending on run environment. (Run-scarce era’s like the late 60’s likely require fewer runs/win.) I’m not sure what value fangraphs uses, but since they’re mainly concentrating on players for whom they have UZR data (2002+), the run environment hasn’t changed much and they can probably use the same value for all players.
Leave Dexter alone! You're lucky he even performs for you!
I'd personally prefer to see a tRA based WAR for pitchers
Mike McCoy Status: FREEEEEEEEEEEEEEEEEEEEEEEEE
Seth Smith Status: Bludgeoning Baseballs
Chris Iannetta Status: OH MY GOD GET SOME HITS YOU'RE MAKING ME LOOK LIKE AN ASS
Check out the most recent MLB Transactions on MLB Daily Dish
by Andrew Martin on Sep 17, 2009 2:39 PM MDT up reply actions
I think it has many of the same issues as FIP
I know we’ve had this conversation before, but tRA has some of the same issues as FIP for a value-determining stat, in that it’s abstracting from the order of events.
Since fangraphs has added tRA and they’re the ones using FIP for pitcher WAR, I think it might make sense to switch. Esp. if their batted ball data is from BIS and matches what is input into their UZR data.
Otherwise, though, I think I prefer Smith’s WAR, or at least would want WAR calculated off both values (off actual runs adjusted for what we can determine about the team defense, and off a run estimator like FIP/tRA).
Leave Dexter alone! You're lucky he even performs for you!
I realize they have the same shortcomings
but Value is a body-of-work type measure as well, so whether those dingers happened before or after the BB, they all count the same, and are written off as normal statistical noise.
Personally, I think they should just use pRAA to determine value.
Mike McCoy Status: FREEEEEEEEEEEEEEEEEEEEEEEEE
Seth Smith Status: Bludgeoning Baseballs
Chris Iannetta Status: OH MY GOD GET SOME HITS YOU'RE MAKING ME LOOK LIKE AN ASS
Check out the most recent MLB Transactions on MLB Daily Dish
by Andrew Martin on Sep 17, 2009 9:41 PM MDT up reply actions
Great comprehensive explanation...
couldn’t have said it better myself.
See Smith’s glossary for further explanation on individual stats.
Eschew Obfuscation!
Rockies Celeberate
in the biggest game of the season, the players stretched in front of the dugout, many happily singing along to Tom Petty’s “Free Fallin’.” From Troy Renck
I think ominous music like this should be banned from the club house for the rest of the season.

by 
















