My Fantasy Baseball Team

Is it any good? What do you think? Before you answer that, I gotta tell you that this is the first time I've participated in this league, and it's a weird one. They use a bunch of weird stats, not the usual 5x5, or even the weird-but-consistent Sabermetric version. They also count 8 hitting stats, but only 6 pitching: R, HR, RBI, SB, BB, FPCT, AVG, OBP, W, SV, K, ERA, WHIP, K/9. I've never drafted a team where you have to worry about fielding percentage, or OBP, or BB, or K/9. Anyway, since OBP and Fielding percentage count in this league and I didn't know what the fuck to do, I drafted guys based on their projected Wins Above Replacement Level, from Baseball Prospectus. So, is my team any good? I have no idea.

Not Quite Happy With That Last One . . .

. . . so here it is again:

This time, I used PECOTA projections for the whole thing. I used Marcel ERAs earlier because I am an idiot and couldn't find the PECOTA ERA projections. Anyway, the results are even more favorable.

Let's assume that all of these guys meet their projections; or at least, any variation one way is cancelled out by some variation the other way. For example, maybe El Duque gets hurt (knock wood) and can't even make his 23 projected starts -- but Perez is slightly better than expected and makes 20 instead of 16. You get the picture.

Anyway, assuming that these five starters can pitch somewhat close to their projection, all they'd need to do to duplicate last year's starters is find 50 starts and 240 innings worth of 5.66 ERA. We'd need to find two guys to make 25 starts a piece, who can give up three runs over four and two-thirds innings. That shouldn't be too hard -- that's like a couple of Steve Trachsels.

The Mets Rotation

Lots of writers are predicting the Phillies, or even the Braves to win the NL East this year; which seems kind of crazy, given that the Mets went to town on the division in 2006 like a fat kid with a cupcake. The 2007 Mets, Braves, and Phillies all ought to have great offenses; but the nobody thinks that the Mets starting rotation will be able to hold up.

Granted, the Mets are relying heavily on two 40-year-olds at the front of the rotation in El Duque and Glavine. On the other hand, they aren't relying on Steve Trachsel and Lima Time, either. I think that the Met rotation is going to surprise a lot of people this season, and could end up, if not a strength exactly, then at least as good as last year. Take a look at this:


Those are a combo of Baseball Prospectus' PECOTA projections for what the Mets starters are going to do this year, along with the Marcel projection for their ERAs. The Mets five starters project to be about the same as last year's starters, or maybe even a little bit better (I'm more bullish on Maine, Pelfrey, and especially Perez than either Marcel or PECOTA, but I digress). Yeah, we'll have to use the Sele's, Parks, and (God help us) Sosas more than I'd like, but there's no way the rotation is going to be as bad as everybody thinks it is.

Something on Batting Order . . . and I’m Back!

This spring, Willie has been messing around with batting Wright second in the order, rather than his usual fifth spot from last year, moving Paul Lo Duca from the 2-spot and dropping him down to sixth or seventh. Is this a good idea? Sure, batting a guy higher up in the order means he’ll get more at-bats over the course of the season, but a lot of managers* subscribe to the idea that each spot in the batting order has a particular job; and some players are much more suited for one job rather than another. I think most players believe this, as well.

What I believe is that batting order doesn’t hardly make any difference at all in how many runs a team scores. If you buy into this kind of thing, which I am sure most major-league managers don’t, David Pinto has got a lineup analysis tool over at Baseball Musings that calculates how many runs you should expect a batting order to score per game. It seems self evident that you’d probably score a different number of runs in a game, even with the same nine guys, if you bat them in a different order.

Anyway, here are two potential lineups for your amusement; one like the standard from last season, with DW batting fifth, and the other with Wright bumped up to second. Included are each hitter’s Marcel** projections for On-base Percentage and Slugging for 2007, which you can plug into Pinto’s lineup analyzer to figure out how many runs each lineup should score. Pitchers are a composite of all Mets pitchers from last season (boy, pitchers are really terrible hitters, aren’t they?) I also included the handedness of the batter, because you know that Willie just loves to do that alternating lefty-righty thing.

What can we conclude from this? It looks like batting Wright second would probably result in about 13 more runs for the Mets than they would get if they batted him fifth. Not a huge difference over the course of a season – it’s less than one run every 10 games. Most of this difference is probably due to the extra plate appearances Wright would get out of the second spot; the number 2-hitter for the Mets came to the plate about 50 more times last season than the number 5-hitter.

On the other hand, most of your smarty-pants sabermetric-types figure that improving a team by 10 runs is worth about one win in the standings (whether by scoring 10 more runs, allowing 10 fewer runs, or by scoring 5 more runs and allowing 5 fewer runs). In what’s probably going to be a much tighter pennant race, with an improved Philly and Atlanta, one extra win in the standings could make the difference between the playoffs and October golf.

I’m going to bet that Willie goes ahead and bats Wright second to start the season. Not because he thinks that giving 50 more plate appearances to Wright (projected .920 OPS) than Paul Lo Duca (projected .744 OPS) would result in more runs, but because he loves that alternating lefty-righty shit. Which just goes to prove that managers can sometimes do the smart thing for stupid reasons.


*Except Billy Martin, who famously drew his batting order out of a hat during a period in the early seventies.

**Marcel is a projection system named after the monkey from Friends, because it’s so simple even a monkey could do it. Despite its simplicity, Marcel is surprisingly accurate.