Wednesday, March 21, 2012

The Path to Respectability: Houston Astros

by Jason Wojciechowski

They're cooked
Photo by Rik Panganiban

Midway through this self-assignment to figure out how each of the crummy MLB teams could be respectable and win 88 games, I've come to my hardest test: the Houston Astros. The Astros aren't a team with a ton of young talent that projection systems don't like but everyone thinks could be good in two years. (They're not the Royals, that is.) They're just a really bad team.

PECOTA sees the Astros having the worst offense in the National League, and it's not even close. Their .243 projected team True Average is nine points below the next contender, the Pirates. For context, if you had two players with those TAvs (you'd have two crap players, but that's an aside), over the course of a season (650 PAs), the Pirates' player would be worth 6.5 more runs than the Astros' guy. A team, of course, amasses a lot more than 650 PAs in a season.

On top of the offense, PECOTA sees the team allowing, through a combination of bad defense and weak pitching, more runs than anyone else in the National League. Including Colorado. Who play on the moon.

But hey! The team has a new owner who's hired a new, more progressive front office. So hope is rising. Is there any hope for this year? Let's get this team to 88 wins.

PECOTA and the BP depth charts have the team at 60-102 (another aside: it's astounding that a team can be projected to over 100 wins), so we need to find 28 wins. As always, I'll lop five wins off of that figure due to good run distribution luck, leaving me needing "only" 23 on the roster itself. Unlike with other teams, there's no picking and choosing here. I'm going to need help from every single player and position.

First, at catcher, we have Jason Castro and Chris Snyder. We could let Snyder start, since he's a better hitter than Castro, but I have a more fun idea: let's suppose that Jason Castro is one of those catchers who actually has a large effect on his pitching staff through pitch-calling, framing, blocking balls, and so forth. The little things, in other words, that until the last year or so, stat-heads, at least in public, have had trouble tracking. We've seen some advances in those areas recently, though, so maybe when we get to the end of the year, we'll find that Castro has shaved ten runs off his pitching staff's totals through all of that activity. That's one win. (For what it's worth, Castro is a Stanford alumnus with a reputation for being a good defensive player, so this isn't entirely implausible.)

Castro can't hit a lick (.243 projected TAv), but he's young enough (25) to learn, so if he can get himself to a little bit above league average, all the way to a .265 TAv, he can add another win with the bat.

At first base, Carlos Lee is 36 years old, but he has a career on-base average of about .340. PECOTA figures him for .326, which, while not unfair (like I said, he's 36), is uncharitable. This, with this solid slugging, works out to a .274 TAv. I need Lee to instead hit .298 (not so far above what he hit last season) and be worth five runs with his glove. (He could instead hit for a .310 TAv and be worth zero with the glove, but I tried to decide which was more reasonable. Yes, Carlos Lee providing positive value with the leather struck me as the more reasonable option. First base isn't that hard!) This adds about 15 runs with the bat and five more with the glove for a two-win improvement. (If he really hit like this, I don't think Brett Wallace would see 25% of the first-base time, but I don't want to futz with that. Pretend Lee gets hurt from trying so hard and Wallace finishes out the last six weeks of the season at first.)

At second base, Jose Altuve is the mightiest of all the tiny men in baseball. (He would be even if David Eckstein were still around.) Still, PECOTA has him with a TAv of just .242, leading to 0.7 WARP. I don't think it's unreasonable to expect him to hit his 80th percentile PECOTA forecast, though, and manage a .300/.332/.416 line. The dude is short, but he destroyed High-A and AA last year. There's no real pressure on the guy, so why not? This would add a win.

At third, Jimmy Paredes is terrible. PECOTA has him worth a full win below replacement, premised mostly on a .270 OBP, in 65% of the third base time. Rather than rescue his performance, I'll instead kick Paredes down to AAA and have Matt Downs play every day. Downs at least has a projected slash line of .252/.316/.408. This comes out to league-average production. If he's the every-day third baseman and adds just five runs to his batting, when combined with removing Paredes from harm's way, the overall increase on the balance sheet is two wins, one for Paredes, half for Downs's bat, and half for Downs's playing time.

(Downs is slated to contribute at second base and right field as well, but with Altuve hitting at second and, as you'll see, a contributor coming forward in right, he won't be needed at those spots any longer. This should take care of any playing-time shuffle issues you're worried about.)

Lastly for the infield, Jed Lowrie was a savvy pickup by new General Manager Jeff Luhnow from the Red Sox, who wanted Mark Melancon for their bullpen. Lowrie has shown flashes of crushing the ball, but he's also only shown flashes of staying healthy. If he can get to 600 PAs this year instead of the 500 he's getting on the depth chart while posting a TAv of about .280 (not quite his 80th percentile -- that's reachable! Plus, he smashed a .319 TAv in 197 great PAs for Boston in 2010), he can add a win and a half to what's currently being expected from shortstop. (When you factor in a playing-time reduction for Marwin Gonzalez, who's basically a replacement-level performer.)

The outfield starts with J.D. Martinez, the team's third-best hitter, in left. PECOTA has him hitting almost exactly what he hit last season (.321 OBP and .424 SLG vs. .319 and .423), but we can put some faith in the demolishings Martinez put on A-ball and AA the last two years and ask him to hit his 70th percentile projection instead: basically .290/.340/.450. That's an eminently reasonable sort of outfielder, a nice league-average kind of guy, and one who's worth half a win more than the depth chart is currently counting on.

In center, Houston has more of a project. Jordan Schafer was a prospect once upon a time, but an HGH suspension and subsequent regression (don't start with me) have resulted in a player who struggles to keep the bat from getting knocked out of his hands. His career on-base percentage is higher than his career slugging. On the other hand, part of the reason for that is a nice walk rate (55 in 532 PAs) and he's flashed some useful speed (24/29 steals despite just 162 hits + walks + HBP (note that this omits ROE and FC)). Combine that with a once-upon-a-time top-prospect profile that came with a good defensive reputation and you've got a recipe for a guy who could be worth 1.5 more wins than the negative 0.4 he's projected for via some combination of his glove and bat. Just hitting .242/.310/.331, hardly stellar numbers, while being +10 in center field would get the job done. Alternatively he could hit something like .255/.330/.350, which is perhaps a taller order, while saving five runs with the leather. Slice it how you like. This team needs Schafer, though thankfully it doesn't need him to perform like a top prospect. It just needs him to be worthy of saying "meh" rather than cringing in horror.

Right field is kind of an unholy mess. The BP depth chart has no fewer than five players getting time out there, but all to no good effect: the total WARP for those players as right-fielders comes out to something like 0.6. That's not pretty. The solution is obvious, at least to someone with my leanings: Jack Cust! He's the best hitter on the team (on-base percentage is king), so asking him to seize the position and run with it for 600 PAs, defense and all, is just asking Brad Mills to recognize game. Now, Cust could hit as PECOTA figures him to (.234/.366/.426), which would be useful, but this team needs more than useful to get to respectability, so I'm going to kick him from a .287 TAv to a .302 TAv (still worse than three of his four seasons in Oakland, for what it's worth) and add two wins to the right-field total.

With the position players finally taken care of, it's time to talk about pitching. Everyone knows Wandy Rodriguez, but it's easy to forget (I could be projecting) that he's already 33, so it might be too late to ask for another three-win season. Still, the 1.4 WARP PECOTA can be bumped up without raising too many eyebrows -- the 1.9 he put up last season in 191 innings looks entirely reachable, and his Fair Run Average hasn't moved more than 0.10 in three years, so we'll pencil him in for 1.9 again, a half-win improvement for Houston.

Livan Hernandez, who I did not remember Houston signing, has a most perplexing projection. Granted that the man is ancient and has been surviving on slop and magic spells for years now, a below-replacement PECOTA when his last three season WARPs have gone 1.8, 3.0, 2.2 is almost aggressively mean-spirited. I don't think it's too much at all, even in front of Houston's defense (check the BABIP projections for the Houston pitching staff), to ask that Hernandez put up 1.0 WARP in 180 innings. That modest production is a 1.5-win increase for the squad.

Bud Norris is a pretty poor excuse for a top home-grown pitcher, but he is what the Astros have, and he's serviceable. PECOTA says 0.6 WARP in 168 innings, but I think a modest goal for the stout righty is a repeat of his 2010 season, in which he posted 1.1 WARP. He actually did that in 15 fewer innings than BP has him throwing, so the quality of his 2012 need not be as good (where "good" means "mediocre") as he showed two years ago. Given that he threw even better in 2011 and that he's only 27, this might be the most reasonable request on the whole list.

In the fourth starter position is J.A. Happ, a barely above replacement lefty who came to Houston in the Roy Oswalt trade with the Phillies. PECOTA says 0.4 WARP on the basis of a 4.5 ERA and too many homers and walks. I say that this is the year for the 29- year-old to find himself through the power of meditation and reach his 90th percentile projection. That'll be worth 0.8 more wins for Houston. (The non-roundness of this number is unfortunate, but it'll work out.)

Jordan Lyles is an incredibly young (didn't turn 21 until after last season) starter, but it's not entirely clear that he's ready. He was about league-average in AAA, which is pretty good for a 20-year-old, but his replacement-level showing in 94 innings at the major-league level showed did not inspire a lot of confidence. PECOTA doesn't see him moving past that easily, so it's going to require yet another 90th percentile performance, an ERA in the low-fours in 180 innings, to add a win of value to his present expected figure.

Finally, there's the bullpen, which has more negative numbers than I've even seen on one depth chart before. Brett Myers is a solid pitcher, and Brandon Lyon is pretty good. Wilton Lopez rates at replacement level. Everyone else, literally everyone, eight pitchers, come out in negative value territory. Three quick fixes for the bullpen, then:

  1. the eight negative value guys average 0.1 WARP apiece. This transforms a unit that is projected to be worth -1.4 WARP into one that's worth 0.8. The 2.2-win upgrade will go a long way for the Astros, and it's feasible. If Sergio Escalona can't hack it, don't let him pitch -- have someone else pick up 0.2 WARP for the both of them.

  2. Brandon Lyon needs to throw twenty more innings -- he's projected for 46, but he's the best reliever the Astros have, so he's got to take the ball. This should add about 0.2 WARP.

  3. Brett Myers has to pitch like he did the last time he was a closer, 2007, when he was worth 1.5 WARP in nearly 70 innings (though 15 of that came in three starts) on the strength of the best FRA of his career. Throwing those 70 innings again and throwing them as well as he did back then would give the Astros the last 1.3 WARP they need to hit the magic number of 88.

(Much of the above comes with some implicit or explicit playing-time increases that cut away some of the chaff: Jordan Schafer's emergence saves 0.3 wins via less playing time for J.B. Shuck; all of Henry Sosa's -0.4 WARP performance disappears due to Livan Hernandez and J.A. Happ; the same for Zach Duke, victimized by Wandy Rodriguez and Jordan Lyles.)

This is going to be the biggest summing up you've ever seen, but here goes anyway:

+2.0 (Jason Castro hits and fields)
+2.0 (Carlos Lee renaissance)
+1.0 (Jose Altuve hits like his minor league performances say he can)
+2.0 (Paredes out at third, Downs in, and Downs hits)
+1.5 (Jed Lowrie beasts, stays healthy)
+0.5 (J.D. Martinez hits average instead of mediocre)
+1.5 (Jordan Schafer gets some swagger back) +2.0 (Jack Cust is a boss, owns right field)
+0.5 (Wandy Rodriguez slows the decline)
+1.5 (Livan Hernandez is ageless)
+0.5 (Bud Norris reaches modest goals)
+0.8 (J.A. Happ meditates, gets good)
+1.0 (Jordan Lyles consolidates, pitches well instead of just young)
+2.2 (entire bullpen is ok instead of being absurdly bad)
+0.2 (Brandon Lyon stays healthy, pitches a fair amount)
+1.3 (Brett Myers reprises 2007)
+1.1 (playing time reductions for Shuck, Sosa, Duke)
+5.0 (run distribution luck)

That's only 87 wins. I did something wrong. I'm pretty sure I had 88. I refuse to fret about this. It's the Astros, and I did my best. They're profoundly not respectable.


Lukehart80 said...

A collision between Joey Votto, Zack Greinke, and Yadier Molina on a pop-up at the All-Star game will simply take the threshold for winning the NL Central down to 87 wins.

Jason Wojciechowski said...

Ha, yes. Leaving 87.1 the perfect number to take this team all the way.

Hugh said...

Your last sentence made me laugh.