3 Up, 3 Down

This post on Fun with Win Shares pointed me towards a Twins' website. In their post Some Sobering Numbers, Three Up Three Down looked at the Twins offseason moves from a Win Shares perspective. Their conclusion: Twins are in trouble.

Well, what about the A's?

The A's won 96 games in 2003, so that means there are 288 Win Shares to distribute. Billy Beane lost/traded players who contributed 105 Win Shares (86 of those WS were Tejada, Foulke, Ramon Hernandez, TLong and Ted Lilly).

That's a lot to replace, but we did get Kotsay, Kielty, Redman, Miller, Hammond and Rhodes. Last year, those six combined for just about 60 Win Shares. So, you might say that we've netted a loss of 45 win shares, and we're headed for an 80-win season. That relatively dire prediction, however, would ignore the potential contributions of our younger players like Rich Harden, Bobby Crosby and Justin Duchscherer. (Clearly the youngsters have limited track records, so projecting out to a full season is terribly risky. But bear with me, I'm being optimistic.)

Here's what I envision for the 2004 A's pitching staff, assuming everyone performs at pretty much the same level. I've highlighted the instances where a pitcher will have to throw more innings. In those cases, I've simply adjusted their win shares accordingly.

Name	2003 Win Shares	2003 IP	2004 IP (Est.)	2004 Win Shares (Est.)
Hudson	23.17	240	240	23.2
Zito	17.66	232	232	17.7
Mulder	16.91	186	200	18.2
Redman	11.15	191	191	11.1
Harden	4.09	75	170	9.3
Duchscherer	1.25	17	80	5.9
Bradford	9.05	77	77	9.1
Hammond	7.04	63	63	7.0
Rincon	6.26	56	56	6.3
Rhodes	4.25	54	54	4.3
Mecir	1.03	37	37	1.0
Harville	0.31	22	22	0.3
Total			1419	113

Considering the 2003 A's pitching staff threw 1440 innings, I feel pretty comfortable about my projections, inning-wise. On to the offense (PA=Plate Appearances, estimated as At Bats + Walks), where I again highlighted differences in playing time.

Name	2003 Win Shares	2003 PA	2004 PA (Est.)	2004 Win Shares (Est.)
Chavez	24.69	650	650	24.7
Ellis	17.82	601	601	17.8
Durazo	16.91	637	637	16.9
Kotsay	13.60	538	600	15.2
Kielty	12.16	498	600	14.7
Hatteberg	14.00	607	607	14.0
Byrnes	15.48	456	350	11.9
Miller	10.11	391	391	10.1
Melhuse	4.07	86	200	9.5
Crosby	0.17	13	600	7.8
McMillon	5.02	172	172	5.0
Dye	1.52	246	400	2.5
Menechino	2.19	102	102	2.2
Total			5910	152

In 2003, the A's had 6,050 plate appearances, so my projection of 5,900 isn't too bad. I should point out, however, that I didn't pay too much attention to getting the right numbr of PA's by position. I'm happy enough that it worked out as it did.

(A couple notes about the playing time adjustments... Kotsay and Kielty, if healthy and given a starting spot all year, should reach 600 PA. Jermaine Dye and Eric Byrnes will end up splitting the remaining OF at-bats. Dye will probably play better than he did in 2003, and Byrnes might play worse. On balanace, their changes will probably cancel each other out in terms of the win shares, so I'm not too concerned about that. --- The one issue of concern is Bobby Crosby. Projecting a full season from only 13 at-bats is ludicrous. But here's why I did it... It works out to only 8 win shares, which would put Crosby as the 15th best shortstop in the AL in 2003. With only fourteen teams in the AL, that would make Crsoby one of the worst every day shortstops in the league. Either he's going to be that good or we'll have to replace him with someone who is, right away.)

So, that's 113 Win Shares for the pitchers, 152 for the offense and a total of 265 Win Shares. After dividing by three, that's 88 wins, but a mere 88 wins isn't going to put anyone into the playoffs. Looking at the last four years, it looks like 91 wins will give you a chance at making the playoffs, and you need 95+ to really feel comfortable about your odds.

I'll leave the remaining 3-10 wins (9-30 Win Shares) in the hands of Billy Beane, and I won't lose any sleep. The A's will be in the playoff hunt all year - well, at least until they clinch a spot...