Jenrry Mejia, The Mets' Bullpen, And The Media

Without naming any names, certain reporters covering the Mets have been expressing their support of the Mets' plan to start Jenrry Mejia in the bullpen and arguing with fans about it on twitter. Watching these arguments unfold, it's painfully clear they don't understand the fan position and think fans are missing something that's painfully obvious to them. Here, I will explain the arguments for Jenrry Mejia NOT starting in the major-league bullpen and how it interacts with the arguments of the beat writers, which at the moment, are not clashing in a constructive way. 

The fundamental premise underlying all arguments for keeping Mejia out of the bullpen is that relievers are inherently less valuable than starters, and by a very significant margin. This gap results from two easily observable realities: relievers pitch less often and the average reliever is much better than the average starter. Your typical set-up man might pitch 60 innings or so in a season, about a third of a typical third starter. Also, the average ERA or equivalent stat for relievers typically hovers around 4.50, while the average starter is closer to 5.50 ERA. Thus, in theory, adequate relievers are much easier to find, even though the Mets have struggled to do so in recent years (but that's another discussion).

Not surprisingly, then, statistics that measure a player's cumulative value to his team over the course of the season rate relievers as a group as much less valuable. Wins Above Replacement (WAR), the most sophisticated statistic in measuring values, has three components: the pitcher's ERA, the ERA of the average AAA replacement he will be measured against, and the number of innings he's pitched. The three most-valuable relievers last season, measured by WAR, were:

• Jonathan Broxton: 2.9
• Matt Thornton: 2.5
• Mike Wuertz: 2.4

By contrast, here are the top three starters:

• Zack Greinke: 9.4
• Justin Verlander: 8.2
• Tim Lincecum: 8.2

It's important to note that these statistics do NOT treat all innings as equal. A leverage factor is individually calculated for each player and factored into the calculation. A closer that pitches often in the 9th or comes in often with runners on base will receive credit for the situation. Now, going back to those top three relievers, here are starters of near equivalent value last season:

• Randy Wolf: 3.0
• Randy Wells: 3.0
• Derek Lowe: 2.7
• Brad Penny: 2.5
• Jon Garland: 2.4
• Barry Zito: 2.2

And for emphasis:

• Mike Pelfrey: 1.8
• Livan Hernandez: 1.7

I will admit that these statistics may underrate relievers a little, as factoring in high-leverage innings is not an exact science. When the best closer, however, represents less than a third of the value of the best starter, clearly there's a disparity. 

This demonstration is not necessarily an argument for Jenrry Mejia making the team as a starter, however. When Mejia starts this season as a reliever, he will be afforded no opportunity to develop his curveball and changeup, essential to his long-term development as a starter (see: Pelfrey, Mike). While the Mets may indicate that the long-term plan remains using him at a starter, this move retards his development as a starter by at least a season. It also starts his arbitration clock, meaning he would begin arbitration and free agency a year earlier than if he just spent two months in the minors. That's potentially two lost seasons of an ace pitching on major-league minimum. And for what? An extra 2 wins in 2010 at absolute most (WAR is scaled to wins, hence the name). Knowing that Kiko Calero would likely assume the 8th inning job in his absence, however, pushing the whole bullpen totem down a notch, and preventing someone like Sean Green from taking the high-leverage situations Mejia would be assuming, the win contribution is likely much less than 2, probably under one win. 

Getting really hypothetical, that's about $1.5 million dollars of value at the current going-rate for a win at the expense of two $15 million dollar value-seasons in the future (assuming Mejia's potential is 5 WAR as a starter). The Mets would be gaining some value now but losing 20-times that in the future. Or in more concrete terms, +1 win this season in exchange for +5 wins each in 2011 and 2012. 

So, members of the sporting press at large, that's the reason we want Mejia to start in the minors. Understanding that, the counter-arguments to anything supporting him in the bullpen should be obvious. But for the sake of discussion...

A set-up man was never the difference between a playoff spot and a 2nd place finish, especially not in this division. The Phillies won the pennant last year with their closer actively trying to blow games. 

First, the Mets bullpen is probably their strong suit this season. Just because it is "unsettled" isn't  a bad thing: it means newcomers are challenging established players with their surprisingly good play. Secondly, no one is arguing that he wouldn't make the bullpen much better, the long-term cost is just staggering. 

Holy cow, if only Mejia only becomes Mariano Rivera...but he doesn't throw a pitch comparable to Mo's cutter, nor does he have the command that allows Rivera to get away with it. Also, Rivera is great, but by nature of being the best of an inherently overrated group, he's always been very overrated. Mejia trying to become Rivera would be a very nice plan B in my mind, if he can't develop secondary pitches. 

False choice. If Green makes the team, he'll be in a mop-up/6th inning role regardless of whether Mejia makes the team, making his performance essentially irrelevant to how the team does. The added value of Mejia is whoever is the next to last currently on the bullpen pecking order (Parnell or Takahashi) minus Sean Green, which is a coin flip. 

Two other names that get compared to Mejia, mostly errantly: 

Joba Chamberlain: The difference between Jenrry Mejia and Joba Chamberlain is the drop-off between Joba's fastball coming out of the bullpen vs. being a starter. If anything, Chamberlain is a reason why starting starter prospects in the bullpen blows up. And on the Mets, Joba would definitely start.

Johan Santana: Johan is a special situation, in that he was a non-prospect, late-bloomer, whose second best pitch, a changeup, was already developed when he was in the bullpen. And because the changeup is such a good weapon against both lefties and righties, he never really needed to worry about his slider being that great. 

I appreciate the beat writers responding to the fans, maybe mostly out of boredom in PSL, but twitter isn't the best place to have an exchange of ideas and hopefully this post prevents future confusion. 

Partly inspired by this LL post.