Monday, February 27, 2006

Early predictions broken down by position

So SuperNova and I had a debate today where we tried to break down each position on the Nats and deduce how many games the Nats would be better or worse with the 2006 version of x fielder/batter/pitcher than the 2005 version. From that we could extrapolate how the Nats would do compared to the 81-81 record of last year. Here's the results:

Catcher: Brian Schneider and Gary Bennett vs. Schneider/LeCroy/Fick.
SNV and I both think that Schneider will take a slight bump down in his offense (and I would say slightly in defense) from last year. The bat of LeCroy/Fick breaks even or a slight upgrade over the loss of Bennett's defense (but see bench rating). Overall, SNV and Dexy both say -1 game.

First Base: Not much comment here, as we both think Nick will stay relatively even. Dexy has 1B as a straight Even while SNV has -0.5 games.

Second base: If Vidro is truly healthy, it goes up more than this, but that is hard to know. This assumes Vidro playing second at moderate-good health. Dexy: +1 game; SNV: +1.5 games.

Shortstop: I have to believe that while Guzman will still be positively awful, he will a) be better than last year and b) they will get someone else in there if he is as bad. Dexy: +2 games; SNV: +1 game.

Third base: This is where we had a big difference. I think that Zimmerman will post an OBP in the .355-.360 range with a little upside above it. Add in a .445-.450 slugging and you get a guy with an .810 OPS (I'd say .800-.830). SNV sees Zim's stats at more like: .330 OBP and .420 SLG. Dexy is being way bullish and saying: + 4 games. SNV says Castilla wasn't horrible and Zim translates into: +1 game.

Left Field: Soriano (if he even plays) vs. Church/Byrd/Wilson platoon: I hate the Soriano signing, but I have this position staying relatively even (i.e. we could have saved $10 million) for this position. SNV says Soriano will definitely be worse than if we had the 2005 platoon in there. Dexy: Even; SNV: -1 game.

Center field: Another big difference of opinion here. I see Ryan Church, IF he gets most of the playing time, as being a slight downgrade from Wilkerson. SNV sees the Church/Byrd/Watson CF platoon as being disasterous compared to Wilky. Dexy: -1 game; SNV: -3 games. So, note that even if Soriano wasn't being paid $10 million, just Wilkerson money, Dexy views the Soriano/Wilkerson trade as costing the Nats 1 game total and SNV sees it costing the Nats 4 games. If Church rides the pine, it gets even more negative for me.

Right Field: SNV and I both worry about Guillen's injuries. Dexy: -2 games; SNV: -1.5 games.

Offensive bench: We did get better in this area, especially now that the lefty-killing LeCroy will see a lot of pinch hitting opportunities. Dexy: +1 game; SNV: +1 game.

Total offense: Dexy: +3 games; SNV: -3.5 games.

Starting pitching:
SP #1: Livan 2005 vs. Livan 2006. SNV sees this as a wash. Unfortunately, I think second half 2005 Livan was more like the pitcher we will see in 2006. Not that bad, but closer to that than first half Livan. Dexy: -2 games; SNV: Even.

SP #2: Patterson vs. Patterson. He's our real ace now. We both think he will come a little back to Earth this year though. Dexy and SNV both: -1 game.

SP #3: Ortiz vs. Loaiza: You go from a guy who was league average, maybe even slightly better, to a guy well below. We are assuming Lawrence is out for the season (or any real Brian Lawrence is). Dexy: -3 games; SNV: -4 games.

SP #4: Ohka/Carrasco/Armas to Armas: Ohka and Carrasco had some nice starts for the Nats. Armas didn't. We now get to see a guy that barely held on to a major league spot last year operate as a #4 pitcher. Dexy: -3 games; SNV: -3.5 games.

SP# 5: Armas/Drese to Drese/Rauch: It's hard to know what will happen here. I suppose there is a chance that Drese could be better, that Rauch will be as good or better than Armas and I am being too negative. That being said, this isn't going to be pretty (and I will be thrilled to apologize if I am wrong). Dexy: -3 games; SNV: -1 game.

Relief pitching. SNV didn't go position by position and says that the bullpen overall will be -1 game worse than last year. I wish I could believe that. With this starting pitching, I see them being overworked even faster this year. I'll say Eichen: Even; Ayala: -1 game; Majewski: -1 game; more use of Stanton: -1 game; Cordero: I hope I am wrong but can see him coming down a bit off his great 2005: -2 games. Total relief prediction for Dexy: -5 games.

After all of that, BOTH Dexy and SNV get to a total of: -13 games. We got there in different ways, with SNV thinking the offense will actually be a bit worse than last year and me thinking that Zimmerman will be a bunch better and the Soriano trade will only hurt a little, but that the bullpen will suffer more this year. We both believe that the majority of the Nats' decline will be due to the starting pitching. Take our 13 game prediction from the 2005 base of 81-81 and you get a 68-94 record assuming, of course, we start the season with the same info we have today.

Here's hoping we are completely wrong.


At 1:23 PM, Blogger Sam said...

Well, you forgot, Frank Robinson is a year older and a year grumpier and a year more stubborn. Given how those qualities were responsible for all the good that came last year* -- +10 games.

**"year" meaning the baseball season that ended on July 4, for which Robinson gets credit. All baseball played after July 4 was Ryan Church's fault for being too much of a softy.

At 1:34 PM, Blogger El Gran Color Naranja said...

Starting pitching is going to be key 1-3 were so good last year in keeping the scores down that our anemic offense could push through a few mroe wins. I think this year the last 2 in the rotation are awful and the first 3 (whoever the third might be) get a lot more variable in their starts. A minor increase in offense isn't going to make up for that. (though I'm thinking more in the 72-75 game range)

At 7:15 PM, Anonymous Will said...

So let's say baseball is 50% offense and 50% defense. And let's say that defense is about 30% fielding and 70% pitching. And 2/3 of the teams innings are pitched by starters. But on average, a reliever's innings are more crucial to the win, let's say 20% more. So an average bullpen will provide about 81*.5*.7*.33*1.2=11 wins. Considering this, -5 wins for our bullpen seems like a huge drop, unless you think our bullpen was really spectacular last year (for the whole season).

By these same guesses, position players are responsible for about 65% of winnage. Assuming this winnage is distributed equally among positions, the average position should yield about 81*.65/8=6.5 wins. This leaves our average starters to provide 17 wins for our hypothetical average team. Predictions for a loss of 12 wins due to starters seem exaggerated.

At 7:26 PM, Anonymous Will said...

Which I suppose gives us another way to predict win totals. Think the Nats' starters will be 20% worse than league average? That's .8*17=13.6 wins from the starters. Repeat for the bullpen and each position and add. If you disagree with my rough breakdown of winnage from the first four sentences of my last post, replace your own guesses and recalculate average win values. Adjust likewise if you think the Nats' current rotation won't be able to get to 6 innings a game. Or if you think that some positions contribute significantly more wins on average than others.

At 7:33 PM, Blogger King said...

You guys are underating Ramon Ortiz's ability to pitch in a pitcher's park. Loaiza was equally bad with some past good years behind him when we signed him.

Astacio should be better than Lawrence if he can avoid becoming old.

At 7:37 AM, Blogger Steven J Berke said...

Not going to argue that much with your position breakdown. But one thing you have to keep in mind in projecting W/L record is that it depends not only on the quality of the team itself but on the quality of the competition.

Last year in the NL East, the Nats faced off with with three pretty good teams (Braves, Phils, Mets) and one decent one (Marlins). This coming year, the Mets may play a little better and the Braves a little worse. The Phillies will be about the same. The Marlins, it is generally agreed, will go from decent to godawful (except of course when Dontrelle is pitching).

So the Nats are unlikely, I agree, to win 81 games again. They, however, may win a few more than you project simply because they get to play this year's Marlins rather than last year's. And for that same reason, they likely will finish fourth rather than last.

At 11:40 AM, Blogger dexys_midnight said...

Responses to Will and Steven:
Will, I appreciate the analysis, but the way you break it down, one would have to go under the assumption that teams a) rarely deviate from 81-81 and b) teams rarely fluctuate from one year to another. But we know that would be wildly inaccurate. Replace last year's Royals with last year's White Sox and you get an increase of 43 wins. Attributing most of that to starters and position players as you do (let's say 39 of the wins to 5 SPs and 8 position players), that averages to every single White Sock being 3 games better than every single Royal. So, considering most of our changes are 3 games or under (and the 3+ game changes are all due to a complete change of player), I find it hard to believe that we exaggerated anything. Do you believe that this team as currently put together will break 70 wins? They only had 67 wins just a year prior (so all we are saying is we anticipate a drop back to that level). Do you see a noticeable improvement from the 04 Expos to the 06 Nats?

Steven, I agree with your point completely and honestly didn't think about it as much as I could have. Although, wouldn't you say that a) the Mets are significantly better this year and b) our facing the worse Marlins is countered by our facing the Yankees/Red Sox/Orioles/Orioles/a much better Toronto/Devil Rays in interleague as opposed to Mariners/A's when they absolutely sucked/Toronto/Toronto/Angels/Rangers? Also, I shudder to think of this, but some of our predictions are best case scenarios..what if the Soriano thing blows up, what if Livan's career begins a huge slide down, what if Guzman still sucks, what if Vidro and Nick aren't healthy, etc.

Listen, we WANT to be wrong on this. This was just our gut reactions to what we think it looks like at this moment.

At 11:44 AM, Blogger dexys_midnight said...

one last point: Will, by designating 6.5 wins per position player for an average team, but only 17 wins for the entire pitching staff, you are effectively saying that each position player is worth TWICE (6.5 to 3.4) what each starting pitcher is worth. Why would you think that? Livan (1065), Patterson (817) and Loaiza (912) each faced way more batters than any position player has plate appearances.

At 6:39 PM, Anonymous Will said...

Thanks for your comments dexys_midnight.

Let's say our bullpen was above-average last year. If they contributed two wins above average, that's 13 wins. If we lose 5 of those wins, it's down to 8 wins. If we can assign our bullpen a winning percentage (this is kinda silly, but it demonstrates the magnitude of the fall), it drops from .591 to .363. This is a huge drop, considering that with only one exception all the main elements of bullpen remain.

Likewise with our starters. If you use my estimate of 17 wins for the average rotation and say that our rotation was really good this year, let's say 21 wins (.618) and you take away twelve wins you get 9 wins by starters (.264). If you think I give starters too little credit and want to give them credit for a third of the wins on an average team, losing twelve wins is still a huge drop from a .574 "winning percentage" to .351 . And two-fifths our rotation is exactly the same, so most of this drop can only be assigned to three players (or less considering that Armas and Drese were in the rotation for good parts of last year). Even if the combined decline of Livan and Patterson is about -3 win (I would guess -1 or -2), then your guesses leave -9 wins for the remaining three starter spots. By your own analysis, we would suffer a drop in all three of these positions equal to the drop between the average player on the World Championship team and the average Royal. I don't remember that our 4 and 5 starters were that good last year, so can they possibly get that much worse?

In response to your second post, I think I addressed both of those issues. I specifically did not distribute those 17 wins equally among the five rotation spots, because this is not even going to be close to true on any team. The front of the rotation starters will have more innings and pitch better and thus get a much greater share of the wins. I distributed the position players wins among positions because I thought the difference between a less valuable position and more valuable one was much less (the offensive differences between positions being almost made up for by defense). Considering that our starting pitchers were better (relative to league average) than our position players, the starters are going to have more than 17 wins to distribute and more of those wins will go to the front of the rotation. But also, in terms of plate appearances, I think it's only fair that some of the credit or responsibility for pitcher's plate appearance go to the defense behind the pitcher.

And yes, I do think this team looks a good bit better than the '04 Expos. Maybe that's just because I'm more familiar with the players and all their upsides, but maybe not. Taking the same assumptions that you guys did, no drastic injuries or collapses, I can see this team easily getting more than 70 wins. I agree with you that the offense will improve and I don't see the pitching in the dire situation that you see it in. I'll go on the record predicting another .500 season.


Post a Comment

<< Home