There’s a school of thought that goes if you repeat a lie loud enough and often enough – it becomes truth. This is, unfortunately, not exclusive to one walk of life. Where it rears its ugly head in baseball is with the notion that you need to have multiple lefty relievers, with at least one of them utilized for only a batter or two at a time, to be successful.
For well over a century, National League baseball thrived without the use of a LOOGY. The first time in NL history that a lefty reliever recorded 30 IP with more games played than innings was 1962. Through 1990, it happened 12 times, with never the same guy doing it twice. From 1991 to 2014, it has occurred a whopping 381 times.
So, what the heck happened?
In 1990, the Oakland A’s won 103 games with two lefty relievers who were right at the LOOGY line. Rick Honeycutt had 63.1 IP in 63 games and Joe Klink had 39.2 IP in 40 games. The A’s were able to do this for a number of reasons. First was an innovative manager in Tony LaRussa who was not afraid to buck conventional wisdom. Also, the team had deep SP, led by Dave Stewart and his 267 IP, which easily led the league. Bob Welch finished third in the league with 238 IP and their fourth SP racked up 199 IP. Finally, they had a shutdown closer in Dennis Eckersley.
The A’s were able to shorten games on both ends, with starters who went deep into games and a closer who essentially guaranteed the game was over after eight innings. It’s easier to play the matchup game when the bullpen doesn’t have to go three or more innings on a regular basis. The following year, the A’s had three lefties in the pen, with Curt Young joining Honeycutt and Klink. It should be noted that the team fell to 84 wins, as the SP did not provide as many innings. Still, the trend was in motion, as six teams employed a lefty pitcher with more games than innings pitched. The following year that number was 17.
And there’s been no looking back since.
The Mets were late to the LOOGY party, not employing their first one until Takashi Kashiwada in 1997. Then came Dennis Cook, who did it four straight years. He was followed by Mark Guthrie in 2002. Mike Stanton then did it in back-to-back years. After a year off in 2005, they came back with Pedro Feliciano and Scott Schoeneweis, leading the way to Tim Byrdak and Scott Rice.
But there was a shift since the Terry Collins/Sandy Alderson regime took over following the 2010 season. The Mets went from an organization that employed LOOGYs to one that developed a blind fealty to the strategy. It made no difference if the lefty relievers were good – they were going to stock them on the roster and use them no matter what. And not only would they use them, they would bend over backwards to get them in a game, even if to do so made no practical sense. The starter cruising with a low pitch count? He’s getting yanked if back-to-back lefties are coming up. Bases empty and down by five runs? If a lefty hitter comes up, a lefty reliever is coming in the game.
The organization was so in love with the idea of employing lefty relievers that it ignored the actual results of the lefties.
From 2011 to 2014, the Mets had 23 lefty seasons in the bullpen turned in by 17 different players. That group combined for 165 ER in 334 innings for a 4.45 ERA. In the same time frame, righty relievers combined for 691 ER in 1,548.1 IP for a 3.93 ERA. The lefty relievers, enjoying shorter stints with more favorable matchups – combined for an ERA 52 points worse than their righty counterparts under the Collins/Anderson regime.
Perhaps ERA is not the best way to measure the effectiveness of the lefty relievers. After all, the idea is for them to come on and retire tough lefty batters in tough situations. Is it possible that they excel in this role but get beat up when they are put in mop-up situations, ones which have little bearing on if the team wins or loses the game?
If this were true, it would certainly be a reason to continue forward with the strategy. Who cares if the Mets lost 10-2 instead of 8-2? It would be a fine tradeoff to make if the lefty relievers were dominating when they came on with a one-run lead and were getting smoked when they entered with the club trailing by five or more runs.
We can use Win Probability Added (WPA) to measure this very thing. The FanGraphs definition of the stat includes this:
“WPA is the ultimate context dependent statistic. You get credit based on how much your action contributes to the odds of winning, meaning a home run in a 1-1 game in the 9th is dramatically more valuable than one in a 10-1 game in the 9th. For this reason, WPA is terrific at telling the story of the game and the players who delivered in big situations. When did the winning team pull away? Who had the decisive hit? These are questions WPA can answer.
It doesn’t tell you how well a player performed, it tells you how important their performance was.”
Last year on April 20, Rice entered a tie game with runners on the corners and one out. He got a double play to end the inning and keep the score the same. Rice earned a .212 WPA for his outing. It wasn’t a long outing, but it was a crucial point of the game and he came on and excelled. Contrast that with his appearance on April 13. He came on in the sixth inning of a game where the Mets were losing, 9-2. In a full inning of work, Rice allowed two runs. That’s an ERA-killing outing but it had no effect on his WPA because the game was out of hand already. He had a 0.00 WPA for this appearance.
Just using these two outings, Rice had an ugly 13.50 ERA but a .212 WPA. Because he performed well when the game was on the line, and rotten when the game was out of hand, his WPA was positive for the two outings. In other words, WPA is the perfect statistic to use to see if the lefty relievers are performing well when it matters and getting lit up when it doesn’t.
In 2014, Rice had a .09 WPA. That’s not particularly good – more on that in a second – but at least it’s better than his 5.93 ERA, which was dreadful. Rice only logged 13.2 innings, so it’s difficult to accurately rate his WPA. But for MLB regulars, a season-long WPA of 1.0 is considered average. Additionally, a WPA of 0.0 is below average and a (-1.0) is considered poor.
Since 2011, the combined 23 seasons of lefty relievers for the Mets have produced a (-1.18) WPA in 334 IP. In my best Bill Walton voice – That’s just terrible! In the statistic that puts the lefty relievers in the best light possible, they’ve been below average in performance. And this isn’t a one-year blip. They’ve been bad over a four-year stretch, despite a manager who gives them the platoon advantage whenever humanly possible.
In the last four years, lefty relievers employed by the Mets have a negative WPA and their ERA is dismal. This despite the fact that Collins runs his bullpen to maximize their value. Yet we’re told over and over and over again how important lefty relievers are to the team’s success. Outside of the manager’s constant bleatings – and the GM’s tacit approval by not putting an end to it – what proof do we have that this is true?
It’s not that the LOOGY strategy can’t work – remember the 1990 A’s – it’s just that it’s not fool-proof and guaranteed to work in all instances. You need starters that go deep into games, a shut-down closer and, perhaps most importantly, good lefty relievers. In the past four years the Mets have not had those things on a consistent basis.
Klink (0.51) and Honeycutt (2.58) had strong WPA numbers for the 1990 A’s because they were good pitchers. In the four years under Collins, only one season has a lefty reliever posted a WPA better than Klink’s 1990 mark. Last year, Josh Edgin had a 0.59 WPA. In a completely unrelated matter, Edgin can actually get RHB out in those rare moments when his manager lets him.
The next time someone tries to tell you that multiple lefties are a necessity, stop them in their tracks and call BS. The only necessity is to have good pitchers. If you have multiple good pitchers who are lefties – by all means stock your bullpen with them. Edgin and Dana Eveland both had good ERAs and positive WPA numbers last year. There was nothing wrong with having them in the bullpen at the same time.
But you are only inviting trouble when you insist on carrying guys solely because they pitch with their left hand. That only makes for a bad bullpen, a point driven home by the 2011 Mets and the 2012 Mets and the 2013 Mets. Last year the Mets had their best bullpen under Collins because the lefties didn’t kill them. With a 3.67 ERA, it’s not like the lefties were great, since the average NL reliever had a 3.53 mark. But the two lefties who saw the most time – Edgin and Eveland – were good relievers.
The game is hard enough; there’s no point in making it harder than it has to be. Take your seven best relievers and if two or more of them are lefties – so be it. You get a good bullpen by collecting good pitchers. You get a rotten bullpen by insisting on carrying guys because they’re lefties regardless of if they’re MLB-quality pitchers. And your manager should not compound the problem by running his pen to benefit the guys who pitch the fewest innings at the expense of the guys who do the most work.
It’s not rocket science.
Good history lesson.
I never thought about why the LOOGY evolved into what it has, and it makes perfect sense why it was created. It was possible when top starters were routinely pumping 225+ innings and even back end ones could reach 200 innings. With an average IP/start of 6.1-6.2 for even the back starters, having a guy who only gets 1-2 outs per game is possible.
But the game has trended once again, and it’s the starters pitching much much less. The top guys reach ~225, and you have back end starters struggling to reach even 180 innings.
Someone has to pick up that slack, and a revolutionary manager is going to win a lot of ballgames if they recognize it’s not optimal to strain 5-6 other guys at the expense the select 1 or 2.
Who was the clown in charge in Oakland that ran Tony Larussa out of town?
Who was the clown in charge of the Padres that ran Bruce Bochy out of town.
Who was the clown that hired and extended Terry Collins as Met manager?
The same guy ran off two great managers at prior stops that have won 6 WS and counting in their careers and hired and extended Terry Collins who has a lifetime .490 winning % and never seen a playoff game except from the stands as a manager to coach our team.
He wants a yes man, he isn’t interested in anyone that thinks for themselves. Thus you get another clown that just throws out a lefty late in the game when he sees someone swinging left handed coming to the plate. If his arm falls out of his shoulder socket, next man up.
“Since 2011, the combined 23 seasons of lefty relievers for the Mets have produced a (-1.18) WPA in 334 IP.” – Brian Joura, putting another nail in the LOOGY coffin.
While that is an interesting correlation Rob, I have never seen anything regarding Alderson “running out of town” Bochy or La Russa. I would love to read info if you have it though.
You’ve got it down to a science at this point Brian. The Mets will hopefully have guys who can go deeper into games this year, but nothing compared to Dave Stewart and the like. Most importantly, the Mets will hopefully just have better quality lefties to choose from, and if one isn’t working, they can bring someone else up quickly.
Once again, Tony La Russa innovated, while everyone else watched in awe and tried to imitate. But here’s a not so subtle secret, he just used guys to the best of their abilities, instead of forcing a particular strategy regardless of his players.
Bochy left to join the Giants with a year left on his contract from a team he had led to consecutive 1st place finishes and the playoffs the two seasons prior to a team that had finished in 3rd 11.5 games behind the Padres the year before. If he wanted a strong willed manager he could have blocked him from making a lateral move to another team or extended him or both. Instead he let him walk without any compensation.
Larussa and Alderson were like oil and vinegar together that is no secret and right after him leaving Alderson made his famous statements about how in no other business was a middle manager put in charge of everything. It basically started when Alderson instituted pretty much the same organizational hitting protocol he has in place now with the Mets and LaRussa told the young kids to upon arriving to the ML’s to “forget that just go out there and hit the best way you know how”. Didn’t endear him to Alderson in the least. But it worked, and that probably stung more than anything.
Collins ineptitude speaks for itself and I can even see hiring him in the first place but to extend him after seeing first hand what he was is inexcusable in my book. But he wouldn’t dare tell a player to disregard Alderson’s organizational hitting philosophy or any other directive even if he thought it was hurting that particular player so there is what is appealing to Alderson imo.
“San Diego Chief Executive Officer Sandy Alderson essentially pushed Bochy out of the job he had held since 1995–the one he moved into as Jim Riggleman was moving out to the Cubs–when he declined to extend Bochy’s contract after San Diego won back-to-back National League West titles in 2005-06.”
“Longtime general manager Kevin Towers and Bochy had been looking over their shoulders since owner John Moores hired Alderson two years ago.”
http://articles.chicagotribune.com/2006-10-29/sports/0610290040_1_general-manager-kevin-towers-bruce-bochy-alderson
LaRussa:
“I knew our programs in Oakland were 100 percent clean,” La Russa told the Daily News in a story published Saturday. “But we had our suspicions — guys hitting stronger but not working out. I went to Sandy and ownership about this. And they told me flat off, ‘Right of privacy. It’s a collective bargaining issue.’ ”
http://www.mercurynews.com/athletics/ci_26225891/tony-la-russa-i-voiced-suspicions-about-players
You think he would have thrown him under the bus like that if there weren’t bad blood there?
My recollection of Dennis Cook was that he was a 7th-8th inning guy in terms of his usage. He happened to be a LHP, but was not a LOOGY (one-out guy) for the Mets.
He wasn’t as extreme as Rice but he still had fewer IP than Games each season with the Mets.
The Cooker I still don’t know how he could see with that pulled down around his nose like that.
Brian, of all the good articles you have written this one was the best. I hope some baseball people in management read it and gain something from it.
Terry Collins is a traditionalist that plays it safe by going by the book and that includes the use of a LOOGY. The last four years it has not mattered if the team won one more game or lost one more game. 73 wins or74 wins, who really cares. This year it may be different though. One more win may mean playing in October or going home in September. Managerial decisions and use of players may be the difference. The use of a LOOGY on this team is detrimental and it hasn’t work. It is one of the reasons that I am dissatisfied with TC as the manager.
Thanks for the kind words!
Let’s see what will happen this season when the Mets don’t have to use any ancient fossils like Farnsworth or Valverde. Can’t wait for TC’s first under the bus when a LOOGY doesn’t execute his match up. As Collins politely points out to writers. i don’t make the pitches. Ah yes! Just one more season for TC and then?
Pete,
I look forward to Harvey, Wright, Flores, Lagares, and all our boys. But, when someone reminds me that Collins is managing this team, I feel like puking.
I just hope they let Harvey pitch. I understand limiting him early to maybe 80-90 pitches to build arm strength. My concern is they limit him to 6 innings to conserve his innings for September and then Collins uses 5 pitchers out of the pen. If only Collins could apply KC’s pen in 2014 as an example I think the Mets have a legitimate shot at a wildcard. Hey Gus we have only one more year of Collins! I hope!
Along with the LOOGY issue, a mirror-image is the Mets failure to utilize a pitcher in a true long man role. I know that Torres is sometimes seen as this type, and he’d be a candidate, but I think teams would benefit by reinstating that role. With the limited innings for Harvey, Wheeler’s struggles to get beyond the 6th, the Mets really need to get pitchers who can go beyond the one-inning role.
Memory not working so well this AM, but there have been many “decent” pitchers who have excelled in that role in the past for the Mets. A lefty long man would be a great addition on a staff loaded with hard-throwing RHP. These days, clubs don’t even seem to look for that kind of player.
Got one: Darren Oliver, 2006. Threw 81 IP in 46 games, went 4-1. Guy was huge for the club that year. Journeyman LHP.
The reverse LOOGY.
It would have been interesting to see if they would have used John Lannan in that role if he had successfully adapted to the pen. My guess is no.
With the way TC uses his pen, where on any given night four or more guys have pitched 4 of the last 5 days, you just can’t carry a guy to pitch multiple innings and then need three days off. It’s all hands on deck, each game, and hope the game doesn’t go extra innings.
Starting pitchers go less and less innings, a trend that likely will continue. Rosters remain at 25. Pitching specialists are a luxury that can no longer be afforded, even when they have indivudual success, the damage to roster flexibility offsets it.