Monday, February 14, 2005

2005 Preview - Houston Astros

I am going to attempt to predict the number of Houston wins this year by the same method that I made my St. Louis prediction. Which is to say, I'm going to look at win shares, OPS+, ERA+, factor in age and injuries, pull some things out of my butt....and presto! Keep in mind when you read this, of course, that I'm not as familiar with the Astros as I am the Cardinals, so some of my points may be even worse than the assumptions I made for St. Louis.


The Astros will head into Spring training with a solid, albeit top-heavy rotation.

1. Roger Clemens
2. Roy Oswalt
3. Andy Pettitte
4. Brandon Backe
5. Pete Munro/Tim Redding/Carlos Herndanez

Note that I'm especially fuzzy about that #5 slot.

So, what can the Astros expect to get this year vs. last?

Roger Clemens posted 20 win shares last year, en route to the Cy Young award. At the age of 41, he went 18-4 with a 2.98 ERA despite pitching in a hitter's park. Impressive stuff. Unfortunately for Astros fans, even if you ignore his age it's highly unlikely that he'll repeat that season in 2005.

His ERA+ in 2004 was a very, very impressive 145 (or 45% better than league average). How good was that? He did not have a season as good in his entire run with the New York Yankees. The last time he had an ERA+ better than that was way back in 1998 when he posted a 176 for the Toronto Blue Jays. In fact, if you look at the last 9 seasons you'll see that Clemens only had 2 years better than 2004 over that time span. The odds of him repeating it again in 2005 (at the age of 42) are slim to none. You see the same thing if you take a look at his win shares.

2004 - 20
2003 - 15
2002 - 11

If you're an optimist, you'll see the trend there and predict that he'll get 25 this year! If you're a realist, you'll probably say that Clemens is more likely to revert to the 15 WS range, costing the Astros about 2 wins that they had last year.

Roy Oswalt had a great season, giving the Astros rotation a fearsome 1-2 punch (which may have caused the Cardinals problems in the NLCS had Phil Garner used them properly in the NLDS.) On the year, Oswalt had 19 win shares and an ERA+ of 123. Interestingly enough, his ERA+ was the worst of his career. However, since he was able to pitch 237 innings, he was able to put together his first 20 win season of his career. (I'm sure he'll do it again a time or two before his career is over.)

With Oswalt only being 27 this season, combined with the fact that he was actually below career numbers last year, this could be a big season for him. Over the last three years, he's posted 20, 10, and 19 win shares, with his 10 WS season coming off of only 127.1 innings. I'm going to predict, however, that Oswalt will "break out" this year (if a 20 game winner can do so) and post 25 win shares, thus making up for any decline in the production of Roger Clemens. Of course, if Oswalt has continued health issues, all bets are off.

Andy Pettitte will give the Astros something they didn't have much of last year - a regular, reliable left handed starter. Look at the ERA+ numbers for Pettitte over the last 5 seasons.

2004 - 111
2003 - 109
2002 - 134
2001 - 112
2000 - 116

Pretty consistent, with that bump in 2002 sticking out as the outlier. If Pettitte stays in the rotation this year, he's likely to produce 12 to 15 win shares for the Astros, which is a step in the right direction for a team that only got 6 out of him last year.

I can only assume that Backe is going to be the #4 starter for the Astros this year, after getting 9 regular season starts in 2004 which were followed by 3 starts in the playoffs. Call me a skeptic, but I think Backe was a flash in the pan. After all - even Jason Simontacchi and Brit Reams had great half-seasons for the Cardinals as starting pitchers, only to crash and burn. Backe had a 100 ERA+ last year, which was the first time in his career he was as good as average.

Last year, Backe posted 5 win shares over just 67 innings. However, I think that it's highly unlikely he can produce at that rate this year. I'm going to take a wild guess and claim that Backe will post 8 win shares in 2005 (which is, in part, based upon the fact that Jeff Suppan and Matt Morris had 7 last year).

Finally, there is the #5 slot. The Astros probably have someone that can fill this slot without killing the team completely. However, that's not a given. Tim Redding, Pete Munro, Carlos Hernandez, and Brandon Duckworth combined to make 51 starts last year, yet only gave the Astros 2 win shares. That's less than 1 win of value, folks. I see more of the same this year. In order to be generous (and allow for any Brandon Backe success stories) I'll give the #5 slot in the Astro rotation 3 win shares this year. Note that this doesn't replace the 7 WS that Wade Miller gave Houston in 2004.


And you thought their rotation was top-heavy? The Astros bullpen posted 39 win shares last year, the same total the Cardinals had. The difference was, however, that Brad Lidge accounted for 43.6% of those win shares all by himself! After Lidge, Dan Micelli posted 7 WS, with Chad Qualls and the departed Octavio Dotel getting 4 each. The rest of the pen gave the team 7 win shares, or just over 2 wins.

Lidge is a stud, plain and simple. However, he's hard to gauge in regard to 2005 performance. He had 17 win shares last year after only posting 8 in 2003. Of course, the fact that he got 29 saves last year vs. 1 the previous season make up for a large number of those WS. His ERA+ also shot up last season, from 123 in 2003 to 227 in 2004. My guess is that Lidge will fall off in ERA+ this season by a bit, but will continue to see his WS value stay high assuming his high workload for the last 2 years doesn't catch up to him. I'm going to predict 20 win shares for Lidge, giving him a value of 1 more win over last season.

The rest of the pen is a large question mark. Dan Micelli is gone (turning Japanese, I really think so), with Dan Wheeler and John Franco coming in. It's interesting to note that the loss of Micelli really might cost the Houston bullpen by a bit, since no one else was really very good at all. What's more, Dotel's production really hasn't been replaced either. Unless, of course, you think that John Franco was just having an off-year last year when he posted an ERA+ of 81 at the age of 43.

Micelli and Dotel combined to provide 11 win shares last year. I'm going to predict that they have only replaced half of that value, or roughly 6 win shares, giving the "non-Lidge" portion of the Houston bullpen a loss of 5. Overall, their bullpen slips by 2 win shares, or almost 1 win.


Here's what the regulars did for Houston last year.

Lance Berkman - 32
Jeff Kent - 23
Jeff Bagwell - 23
Carlos Beltran - 18
Craig Biggio - 18
Adam Everett - 12
Morgan Ensberg - 10
Brad Ausmus - 6

A few items of note from this list. Jeff Kent is gone, which leaves the Astros looking for 8 wins at his position alone. Carlos Beltran is gone, giving them another 6 win loss (which is amazing considering he was only with the team for half of the season.) Not listed are Mike Lamb, Jose Vizcaino, and Richard Hidalgo, who combined to provide another 24 win shares for the Astros as part time starters during the season.

Lance Berkman will likely be the best hitter on the team again this year, which he has been 3 of the last 4. Unfortunately for Mr. Berkman, he'll start the season on the DL. Last year, he had 32 win shares after posting 25 in 2003 and 30 in 2002. He also posted a 161 OPS+ last year, which was his best season since 2001. You can probably expect a drop-off in production from Lance this year, both due to him having a bit of a peak last year, plus reduced playing time. I'm going to project him to have 25 WS in 2005.

Jeff Bagwell is an interesting case. Look at his OPS+ numbers over the past several years.

1999 - 169
2000 - 152
2001 - 141
2002 - 137
2003 - 127
2004 - 117

If that keep up, his 2005 production is going to be about 7 to 10% above league average. Over the past 3 years, Bagwell has posted 23, 22, and 23 win shares. I suspect that he'll have a hard time posting 23 again this year unless he gets an age 37 bump in hitting. Peg him for 20 win shares.

Craig Biggio is another aging "Killer B." Last year he posted an OPS+ of 106, which was actually just the 2nd time in 5 years that he's been above average. He turned 39 on December 14th - think he'll be above average this year? Unlikely. He has had 18, 20, and 15 win shares over the past 3 years. I'll give him the benefit of the doubt and say it will only slip to 15. (Not that you really want a 15 WS player in the corner outfield, mind you.)

The rest of the Houston offense is going to rely on the likes of Morgan Ensberg, Adam Everett, and Jason Lane.

Ensberg had a disappointing season after looking good in 2003. Unfortunately, his 2003 season looks like a fluke if you consider the two seasons on either side. Still, look for Ensberg to provide 12 WS this year over 10 in 2005 if he gets more playing time.

Everett had 12 win shares in 2004 after posting 11 in 2003. Let's keep up the trend and give him 13 this year.

Jason Lane only got 136 at-bats yet still managed to post 6 win shares. If he can get 500 at-bats and keep up that rate? He'll be worth about 20 win shares. That may be a stretch, but I think he can do it.

The Catcher is Brad Ausmus, who has posted 6, 12, and 9 WS over the past 3 seasons. Looks like we can pencil him in for 9.

The 2nd baseman is likely to be Chris Burke, who hit .315/.396/.507 last year in AAA. The strange thing about him is that his 903 OPS in AAA was the first time he's exceeded the 767 mark as a pro. Had he played in pitchers parks? Did he mature? Or was he lucky? I have no idea. All I can say is that his bat is highly unlikely to replace Kent this year, while his glove has to be better. Let's be somewhat optimistic and say that he'll cost the Astros 3 wins from Kent, giving him 14 win shares on the season.


As was the case with the Cardinals, the bench is hard to gauge. Who was really on the bench, and who really was a part time starter? Take Mike Lamb for example. He started in the place of Morgan Ensberg when he was hurt - and proceeded to provide more value to the team (12 WS vs. 10).

Overall, the Astros bench had 36 win shares last year. Mike Lamb, Jose Vizcaino, Orlando Palmeiro, Erik Bruntlett, and Raul Chavez all return, and may be joined by the likes of Charles Gipson and prospect Willy Tavarez. I see no reason as to why the bench is going to be any worse, nor any better, than last year. So, for simplicity sake, I'm going to say that the Astro bench will remain neutral this year, giving them 36 win shares once again.

Grand Total

Starting Rotation - 66 Win Shares (60 in 2004)
Bullpen - 37 Win Shares (39 in 2004)

Starting Lineup - 128 Win Shares (140 in 2004)
Bench - 36 Win Shares (36 in 2004)

That's a grand total of 267 win shares, or 89 wins. From a Pythagorean standpoint, they had 1 "lucky" win in 2004, so we need to take that adjustment here as well, putting my prediction at 88 wins. Does this pass the sanity check?

I'm predicting a Houston rotation better than the 2004 version by 2 wins. This prediction includes Roger Clemens not being as good, Roy Oswalt being a little better, and a full season of Andy Pettitte. It also assumes Brandon Backe being average or slightly below on the year, and the #5 starter being slightly better than last year. It seems high, but within the margin of error.

I'm predicting a bullpen to be a downgrade over 2004 by about 1 win. This takes into account a full year of Lidge as the closer, but with the rest of the pen weaker with Micelli gone, and not getting the benefit of Dotel for half of the year. Once again, this seems like a conservative estimate, but not insane.

The lineup prediction is of course where most of the loss is taken. Berkman out for a month or more, Kent and Beltran gone, Biggio and Bagwell older. The only way my prediction of 4 losses due to those players holds is if in fact Jason Lane and Chris Burke can play at a decent level this season. If either of them do not produce, or if Biggio and/or Bagwell regress more than I'm projecting then the Astros could have trouble playing .500 ball this year.

And, of course, the bench prediction is hard to deal with, but does not seem out of line. If anything, I'm probably off by 1 win in either direction.

So as it stands right now, the Astros are looking like a Wild Card team at best. If they can add some production at either 2nd base or corner outfield (thus moving Biggio back to 2nd), it would be good for their win total. Additionally, the Astros should be on the market for a middle reliever or two, as they really need to bridge the gap between the rotation and their closer.


Post a Comment

<< Home