Build a better mousetrap and the world will beat a path to your door.
This time honored saying, paraphrasing a quote from Emerson, speaks to the power of innovation. Be new, different and better and the world will love you. As Mets fans, we would be thrilled if the team achieved even two of those three things. Meatloaf and all, you know.
Unfortunately, we have to settle for one thing. Old dog Terry Collins isn’t much on new tricks but we should give him credit in trying something different, that being how he chooses to deploy Carlos Torres. In his usage of Torres, Collins seems to be channeling 1974 Mike Marshall. That year the Dodgers used Marshall 106 times and he threw 208.1 innings.
At least Marshall had some build up to that amazing season. It was the third straight year he led the NL in Games and his totals went from 65 to 92 to 106. His innings showed a similar progression, going from 116 to 179 to 208.1 over that span.
Meanwhile, Torres is on pace to appear in 85 games, which would be the fourth-most ever by a Mets pitcher and the most by a righty. He’s also on pace for 97.2 IP, which would be the most by a team pitcher who functioned almost exclusively as a reliever since Roger McDowell tossed 128 innings in 1986.
My opinion is that Collins should be using his relievers in longer stretches. The problem is that he continues to use Torres in a variety of different roles. He has 12 appearances where he’s logged more than an inning of work and nine where he pitched 0.2 innings or fewer.
Collins is using Torres virtually every chance he gets. His usage pattern mimics that of a LOOGY, except that Torres isn’t having one and two-batter appearances. Only five times in 35 appearances has Torres had a LOOGY-like workload. Overall, he’s faced 182 batters in his 40.2 IP.
But it’s more than the Mets and more than just LOOGYs. This century there have been 44 seasons in which a righty has appeared in 80 or more games. Only six pitchers have done it more than once. For every pitcher you can name like Scott Proctor or Paul Quantrill to do it and continue to pitch effectively, you can name guys like Jim Brower, Billy Koch, Geoff Geary and Oscar Villareal who did it and were never the same.
Undoubtedly, part of this is just the nature of the job. There just aren’t a ton of relievers who do the job well for extended stretches. Most are failed starters and if they failed once we shouldn’t be surprised if they fail again.
More than a few of these guys who appeared in 80+ games were being used that often because they were so effective. Of the 44 pitchers in this sample, 32 finished the season with an ERA+ of 120 or above. If you know they’re not likely to last long anyway, you can make an argument that you should get all the value you can while you can.
Currently, Torres has an ERA+ of 98.
Using Torres four times a week is not working in the current season and increases the likelihood of him not being worthwhile in future seasons. We need to do better than the current lose-lose scenario with a guy who has the potential to be so much better than he is.
Imagine if the Mets ran their bullpen to optimize the effectiveness of Torres rather than the LOOGY du joor. Imagine that if instead of saying – we need to make sure our lefty reliever gets used for one or two batters only, regardless of the state of the game or the state of the rest of the relievers – we instead said that Torres was only going to be used twice a week for two or three innings at a time.
Wouldn’t it be better to build your bullpen around one guy pitching four-to-six innings a week rather than building it around a guy pitching to four-to-six batters per week? Instead of saying the LOOGY is going to only pitch to LHB and all other pitchers have to work around his limitations – instead we’ll say that we’re going to optimize the usage of the guy who can pitch to both righties and lefties and go multiple innings at a time.
Then imagine saying that along with Torres, we’re going to use the same approach with Gonzalez Germen and Jeurys Familia, too. So, instead of catering to one guy so that he can pitch to six batters in a week, we’re going to cater to three guys to throw 15 innings a week.
At that point, you add in your closer, looking to get him three innings a week.
The other three relievers – and this includes your LOOGY, if you still insist on carrying one – have to work around the needs of the main relievers. So, instead of having multiple relievers constantly having pitched in four of the past six games, you have a consistent rotation where your workhorses get rest, not because they desperately need it but because it’s built into the system.
There’s about 26 weeks in the baseball season. If Germen, Familia and Torres each pitched twice a week and gave an average of five innings pitched in those appearances, they would be on pace to finish the year with 390 IP, without the constant worry of being burned out like the current system has them.
Of course, things are never that easy. There’s no guarantee that three relievers can stay healthy and effective over an entire season. And you know there are going to be stretches where the team plays back-to-back 14 inning games or ones where the starters go more (or fewer) than six innings per night.
You have to be flexible. But wouldn’t it be nice if there was a week where there were no extra-inning games, no starters got knocked out early and the bullpen could not tax its main guys? Whatever plan you use should have times that are, for lack of a better word, normal.
Our current deployment never has that because it’s constantly struggling to make up for the one and two-batter performances.
From May 4 to May 11, the Mets played seven games and six times the starters went at least six innings. It wasn’t perfect because there were two extra-inning games but it still serves as a good example of the current deployment.
In that seven day stretch, the Mets went to the bullpen 25 times and their relievers supplied 22.1 innings. Kyle Farnsworth and Torres each pitched four times while Familia and Rice each pitched five times. The Mets went 2-5 in this stretch.
Going into the game on May 12, the Mets’ bullpen should have been in decent shape. Instead, Torres and Farnsworth had both pitched in two of the previous three games, Rice had pitched in the previous three and Daisuke Matsuzaka had pitched two innings the previous game. There are four relievers who shouldn’t have pitched after a good week from the starters
Let’s go back and try to work a different bullpen arrangement.
Both Farnsworth and Matsuzaka pitched on May 3, so let’s call them unavailable for the following day. What if the Mets had used this deployment, instead:
5/4 – Familia (2), Valverde (1)
5/5 – Torres (2)
5/6 – Matsuzaka (1)
5/7 – Familia (2), Farnsworth (0.1)
5/8 – Off day
5/9 – Matsuzaka (1.1), Torres (2.0), Farnsworth (1.0), Valverde (1.0), Rice (1.0)
5/10 – Familia (2.0), Farnsworth (1.0)
5/11 – Matsuzaka (2.0), Valverde (1.0), Torres (2.0)
Using a more sensible bullpen deployment, only Torres and Matsuzaka should be out for the game on 5/12, although the preference would be to not use Familia, too. This week was challenging because two of the last three games went extra innings and the Mets used essentially a six-man pen with Germen only pitching 0.1 IP in real life. If the Mets simply called up someone else for me to use instead of Torres on 5/11, it would be much better.
Plus, this was the week that Collins pulled Dillon Gee early to get Rice into the game. The bullpen should have had less work than it did in the 5/10 game.
Still, the Mets should have been able to get by with 16 pitching changes, rather than 25. Now imagine a week where the starters didn’t consistently pitch deep into the game and imagine how many pitching changes a sensible deployment could save. And imagine how well-rested the pen could have been this week if Germen was available and Rice could pitch like a real reliever.
We have no idea if this alternate bullpen deployment would work in 2014. What we do know is that the current model isn’t working and is more likely than not to burn out Torres in the process.
Perhaps with all of the young pitching talent bubbling in the pipeline, the Mets should not be concerned with burning out Torres. Personally, I don’t agree with that statement but at least it makes some type of ruthless sense. But if the goal is to maximize Torres’ effectiveness in the 2014 campaign, the current model which has him overworked, punching himself in the dugout and carrying a 98 ERA+ has been a dismal failure.