Thanks to everyone for the great response to the Red Sox projection thread, which even got a lead-panel mention in Soxaholix (sorry, Deadspin, but that's the coolest thing ever). To simplify all the conclusions in that post, a combination of the two projections and the Jamesian Pythagorean theorum gave the Red Sox the possibility of 111, 105 and 99 wins. I like the 99-63 record best because it simply adds the wins gained from the projected additional runs scored and additional runs prevented to last year's record, which already was affected by the use of bench players and poorer relievers. It's not perfect, but this method at least counts for that performance by assuming it stays unchanged (and for the Sox, it can only get better, right?).
I won't use the Bill James projections for the Yankees (buy the book for yourself!) because we determined that PECOTA was nearly identical on offense and far more realistic on pitching. Otherwise, I'lll do the projections the exact same way, with the same non-bench flaws, so we can have an apples-to-apples comparison.
First of all, the 2006 Yankees' typical starting lineup:
Damon -- .285/.359/.482 24 HR, 169 H, 115 R, 25 SB
Jeter -- .343/.417/.483 14 HR, 97 RBI, 214 H, 34 SB
Abreu* -- .297/.424/.462 15 HR, 107 RBI, 41 2B
Rodriguez -- .290/.392/.523 35 HR, 121 RBI, 26 2B, 15 SB
Giambi -- .253/.413/.558 37 HR, 113 RBI, 25 2B
Posada -- .277/.374/.492 23 HR, 93 RBI
Cano -- .342/.365/.525 15 HR, 78 RBI, 41 2B
Williams -- .281/.332/.436 12 HR, 61 RBI, 29 2B
Cabrera -- .280/.360/.391 7 HR, 50 RBI, 26 2B
*Includes numbers with Philadelphia.
The lineup scored 930 runs, or 5.74 per game. It should have scored about 6.33 runs per game, according to the Lineup Analysis, or an astounding 1,025 runs. That's seven lost wins. Of course, Abreu wasn't there until August, and this lineup was by no means the most used. Andy Phillips and Miguel Cairo had significant playing time during the year. Like the Red Sox, injuries helped drag these numbers down. As everyone said toward the end of the year, replace Williams and Cabrera with Sheffield and Matsui, and this is a prolific -- even historic -- lineup. Thankfully, from my perspective, they never all got healthy at the same time.
So how does the 2007 lineup project, according to PECOTA?
Damon -- .289/.362/.458 18 HR, 158 H, 94 R, 14 SB
Jeter -- .322/.390/.452 12 HR, 71 RBI, 189 H, 23 SB
Abreu -- .277/.389/.447 16 HR, 65 RBI, 28 2B
Rodriguez--.288/.385/.531 34 HR, 104 RBI, 30 2B, 14 SB
Giambi -- .252/.413/.518 29 HR, 81 RBI, 18 2B
Matsui -- .288/.376/.474 15 HR, 60 RBI, 21 2B
Posada -- .259/.365/.433 17 HR, 65 RBI
Cano -- .308/.345/.472 16 HR, 80 RBI, 36 2B
Mientkie -- .251/.328/.382 5 HR, 30 RBI
PECOTA projects dropoffs from every single member of the Yankee lineup except A-Rod, who projects to be basically the same. Damon's power will drop, Jeter's averages will drop by 20 points, Abreu will lose 20 to 30 points off all his averages, Giambi's power will drop, Posada drops by 20 points, Cano's counting stats go up while his rate stats decline, Matsui only gets 361 ABs and Mientkewicz becomes the rivalry's next Alex Gonzalez.
The Yankees' lineup is aging, so drops for most of its players seems realistic, particularly when you factor in the hugeness of the years Jeter and Cano had in '06. I'm surprised PECOTA is so down on Matsui and Abreu -- particularly considering Abreu raked with New York; it was his numbers in the NL that held him down. I also would have thought they would project an uptick from Rodriguez. If the Yankees go from having two 35-homer guys and four 20-homer guys to just one 30-homer guy and two 20-homer guys -- and from seven players within 20 points of .300 to just five -- I'd be pretty happy as a Sox fan.
Nevertheless, this is still an excellent lineup, one of the best -- if not the best -- in baseball. According to the lineup tool, it should score 5.91 runs per game, or 957 runs total. That's 27 runs better than how they performed last season, but 68 runs worse than how they should have performed. Is the glass half empty or half full? This is a different situation because with the Red Sox, their projected RPG was better then both their 2006 RPGs (the one they actually had, and the one they should have had based on their stats). To arrive at the 99-win total for the Sox, we used the '06 reality, so let's do that here. The Yanks gain three wins -- essentially by having Abreu the whole season.
Mussina -- 15-7, 3.51 ERA, 1.10 WHIP, 35 BB, 172 Ks
Wang -- 19-6, 3.63 ERA, 1.31 WHIP, 52 BB, 76 Ks
Johnson -- 17-11, 5.00 ERA, 1.24 WHIP, 60 BB, 172 Ks
Wright -- 11-7, 4.49 ERA, 1.53 WHIP, 57 BB, 84 Ks
Chacon -- 5-3, 7.00 ERA, 1.79 WHIP, 36 BB, 35 Ks
Lidle -- 4-3, 5.16 ERA, 1.50 WHIP, 19 BB, 32 Ks
Now the bullpen's top arms
Rivera -- 5-5, 1.80 ERA, 0.96 WHIP, 34 Sv, 11 BB, 55 Ks
Farnsworth--3-6, 4.36 ERA, 1.36 WHIP, 6 Sv, 28 BB, 75 Ks
Proctor -- 6-4, 3.52 ERA, 1.19 WHIP, 33 BB, 89 Ks
Villone -- 3-3, 5.04 ERA, 1.57 WHIP, 51 BB, 72 Ks
Overall, the Yanks' staff allowed 767 runs.
Here's the PECOTA projections:
Mussina -- 12-9, 4.27 ERA, 1.31 WHIP, 47 BB, 146 Ks
Wang -- 11-9, 4.35 ERA, 1.42 WHIP, 48 BB, 74 Ks
Pettitte -- 12-9, 4.21 ERA, 1.34 WHIP, 51 BB, 131 Ks
Pavano -- 6-6, 4.64 ERA, 1.39 WHIP, 22 BB, 54 Ks (15 GS)
Igawa -- 11-9, 4.46 ERA, 1.40 WHIP, 59 BB, 128 Ks
Karstens -- 7-9, 5.55 ERA, 1.49 WHIP, 46 BB, 81 Ks
PECOTA doesn't like the Yanks' rotation either, projecting declines for Mussina and Wang, a solid season from Pettitte, about what could be expected from Igawa. Karstens doesn't look good at all.
And the PECOTA bullpen:
Rivera -- 4-4, 2.78 ERA, 1.18 WHIP, 27 Sv, 13 BB, 47 Ks
Farnsworth--3-3, 3.85 ERA, 1.34 WHIP, 5 Sv, 27 BB, 44 Ks
Proctor -- 3-3, 4.39 ERA, 1.37 WHIP, 22 BB, 50 Ks
Villone -- 3-2, 4.90 ERA, 1.53 WHIP, 27 BB, 45 Ks
Vizcaino -- 3-3, 3.89 ERA, 1.40 WHIP, 23 BB, 49 Ks
I hope I got the bulpen right. Not being that versed in the intricacies of the Yankee hot stove, I might have missed a middle relief acquisition somewhere. As it is, I'm just trying to do the four or five likely to see the most innings. Knowing Torre, I bet he'll find some time in there for Villone and Proctor ;-).
At any rate, these numbers extrapolated over the 1,444 innings the Yankees played in 2006 gives them 738 runs allowed, 29 runs better than in 2006. As was mentioned on the Sox projection, this doesn't account for the myriad relievers and fill-in starters over the course of 2007 who may or may not perform better than the 11 listed here.
So, the Yankees improved by three wins on offense and three wins on pitching, giving them six total wins of improvement. By that measure, the Yankees would finish at 103-59.
Using the same Pythagorean formula as I did on the Red Sox, the Yankees' projected 2007 of 957 runs scored and 738 runs allowed would come to a 102-60 record.
Wins added method:
Red Sox: 99-63
Red Sox: 105-57
Finally, Nate Silver at Baseball Prospectus, which releases the PECOTA projections, must have been snooping around YFSF because out of nowhere the other day -- well after I'd posted the Sox projections -- he ran the PECOTA runs scored and allowed projections for the Sox and Yanks, and only the Sox and Yanks:
Just a very preliminary attempt to spit out team projections for the Yankees and the Red Sox based on the PECOTAs.RS RA
Yankees 918 774
Red Sox 913 772
This is going to be fun. Well, provided that you live north of the Mason-Dixon line and
westeast of the Appalachians.
Hopefully PECOTA's projections are a little better than its geography :-P. That's 93 more runs on offense for the Red Sox and and one fewer run on the pitching side. For the Yankees, it's 11 fewer runs on offense and eight more runs given up. That means the Sox gain nine wins over last season while the Yanks lose two. That puts the records like this:
Red Sox: 95-67
Using Jamesagoras and those totals, we get:
Yankees: 95-67 (Actually 94.7)
Red Sox: 94-68 (Actually 94.46)
Looks like another wild ride.