There seem to be two schools of thought when settling on a 25-man roster: Either the Mets' entire season rides on the last man on the bench and in the bullpen, or the decisions made will have absolutely no impact whatsoever.
The truth -- as it so often does -- lies somewhere in the middle. We'll take a look at recent battles to get an accurate sense of this. Yet it is hard to shake the feeling that the 2010 choices say something larger about this team and its inability to get out of its own way.
Way back in 2009, the final bench spot came down to signing Gary Sheffield over Marlon Anderson. Needless to say, this decision had a large impact on making what was a lost season slightly more bearable. While Sheffield posted an .823 OPS in 312 at-bats, and Anderson failed to crack a .600 OPS -- with Newark of the Atlantic League. In this case, a player the Mets thought deserved a multi-year, Major League guarantee was deemed unworthy of a spot in any other organization.
Then again, other decisions in 2009 had less of an impact. Jeremy Reed over Cory Sullivan meant little; the two went on to have nearly the same number of plate appearances. (Fun fact: Reed impressed in the spring, but Sullivan's .720 OPS far outpaced Reed's .605 mark. Again, Spring Training performances are meaningless.)
Take a step back one year to 2008, and the final bench spot mattered quite a bit. The Mets went with Fernando Tatis over Ruben Gotay and clearly made the right call. Tatis went out and posted an .853 OPS over 306 plate appearances. Angel Pagan also got the chance to impress with the final outfield spot, and Joe Smith got to keep Duaner Sanchez's bullpen spot warm. So two out of three ain't bad.
Of course, some decisions can go the other way, too. In 2007, the Mets chose to keep Ambiorix Burgos over Chan Ho Park, since they didn't know how Park would profile as a reliever. Park has gone on to perform very nicely as a reliever with the Phillies and Dodgers (Sunday night's Yankees debut notwithstanding). And in 2006, the Mets kept Victor Diaz over an extra pitcher named Heath Bell.
Think the 2006 NLCS might have turned out differently with Bell and not Guillermo Mota pitching in key spots? How about the 2007 and 2008 stretch runs with Bell and Park out of the bullpen?
For what it's worth, I blame the Mets for Bell but not Park, whose second act seems like the best evidence yet that F. Scott Fitzgerald, in this case, was way off.
The point here is not to bury the Mets for having hindsight in advance. Rather, it is that thinking carefully about these roster spots is more than just a theoretical exercise.
So what do we take from the decision to go with Sean Green, Fernando Nieve and Jenrry Mejia ahead of Nelson Figueroa and Bobby Parnell in the bullpen, and Mike Jacobs and Gary Matthews Jr. ahead of Chris Carter and Angel Pagan in the starting lineup?
Green is possibly the most glaring of the poor judgments, even ahead of Mejia, because while the Mejia decision simply looms as the harbinger of damage they'll do in the future, Green is damage done now.
Let's get this straight: Sean Green was not an effective member of the bullpen in 2009. He walked 4.7 batters per nine innings while he continued his career-long trend of failing to get lefties out. This winter, he reworked his pitching motion, dropping down more (which should help lefties see the ball even better out of his hand) and walked 6.9 batters per nine innings.
Spring Training stats don't matter, but when they simply reinforce the flaws of a pitcher that were already accumulated over several hundred innings, they certainly can't serve as the positive reinforcement necessary to make the Mets give Green another Major League shot.
In other words, Sean Green is on this team because he makes $925,000 in 2010, and the Mets don't realize that a sunk cost means they can pay him $925K to try and figure it out at Buffalo, or he can actively harm the Major League roster.
Meanwhile, Bobby Parnell displayed no platoon split last season (OPS against lefties: .739, against righties: .766), is younger, had a better strikeout rate, better walk rate and was significantly better as a reliever in 2009 than Sean Green did. And he's at Triple-A.
So is Nelson Figueroa, in the best-case scenario for the Mets. Figueroa can start or relieve, and with no shortage of question marks on the staff -- both in the rotation and with the three relievers who have yet to throw a Major League pitch -- Figueroa would be useful insurance. Instead, he's likely to bolt for Japan, and wouldn't you?
Fun fact: none of the three question marks in the bullpen who have yet to throw a Major League pitch is the aforementioned Green or Fernando Nieve, who has 47 1/3 career Major League innings in relief.
And not to put too fine a point on it, but in starting Mike Jacobs at first base and Gary Matthews Jr. in center field, the Mets are choosing experience over talent. Chris Carter and Angel Pagan may or may not be effective Major League regulars. But we know that Jacobs and Matthews Jr. aren't.
So with first pitch just hours away, my customary optimism has been tempered quite a bit. If the Mets won't see Green and his $925,000 as a sunk cost, what will it take to see Luis Castillo's $6 million salary as a sunk cost? Probably a new GM, you'd have to figure.
It is still a roster with intense variability. If this team wins 95 games, it wouldn't surprise me, with the star power and upside of secondary players. If this team wins 65 games, it wouldn't surprise me, with the injuries to star players and low floor of secondary players.
Unfortunately, the answer probably lies somewhere in the middle. And the choices made for those final roster spots do not inspire confidence that the fine-tuning necessary to get those extra wins will be adept enough to do it.