I really hate to make my first comment on BtB be about this, but I just can’t resist. I have a few issues with the idea that this year’s power rankings would be based on a run adjustment for hitters and pitchers that can approximately explain the AL’s win percentage over the last 5 years.

First off, does anyone really think that the performance of the NL players today has anything at all to do with it their performance in 2004? 5 years is a long time in baseball. People get traded, youngsters come up, and people get older. I don’t have to time to look it up, but I’d venture a guess that less than 1/2 the players who started 2004 on an NL 25 man roster are still on one. And even if they are, they are 5 years older. No, counting back to 2004 does not make much sense at all. Additionally, what’s the predictive value of looking 5 years back? We don’t have alot of data, but we can look at what we do have. Starting in 2002 this is the AL win % vs their win percentage that previous 5 years (what you would have used to create the power rankings that year).
              AL Win % / AL Win % Pre 5 yrs / Difference
2002 48.81% / 49.92 % / 1.11 %
2003 45.63% / 50.49 % / 4.85 %
2004 50.20% / 49.44% / 0.76 %
2005 53.97% / 50.24% / 3.73%
2006 61.11% / 50.20% / 10.91%
2007 54.37% / 51.95% / 2.42%
2008 59.13% / 53.06% / 6.07%

Average variance is 4.26%. FWIW, if your prediction was set at 50% each year instead of the last 5 year win%, the variance would only be 4.9%. So it doesn’t seem to be a very good model does it? Likewise, just take last years win % doesn’t seem to predict next win % all that well either. Every year, the leagues are different than they were the previous year, we should expect variations.

I have to run out, but suffice it to say that is just one of my issues with the AL vs NL adjustments. The others center around the fact that the AL and NL don’t play the same game, even if they do call it baseball in both leagues.

I am a little familiar. I get it, pitchers do worse moving from NL to AL, as do hitters (though I haven’t seen as much on the hitters). But once again I would say that has to do with the AL and NL not being the same game. There are plenty of intangibles ( or rather, hard to calculate) that go in to having the DH other than just adding an additional bat in the lineup. I realize specific examples won’t prove anything, but none the less look at the Cardinals this year. Carpenter goes down swinging the bat, misses a month. Lohse get hit by pitch, twice this season, this time it’s going to cost him at least 2 starts. More generally, pitchers have to hit, run the bases, then come back and pitch the next inning rather than sit comfortably sipping gatorade. The double switch takes ABs away from starters, more replacement / bench types get in games in the NL.

What about injury prone guys in the field? Milton Bradley tears in up in the AL last year as DH, looks awful this year as a Cubbie. In less extreme examples, what’s the effect of being able to “rest” Vlad Guerrero in the DH spot every couple of days? Does it make him that much sharper in the field when he plays there, being “fresher”. And not having to actually sit him for a day, does he keep his timing a little better at the plate?

Then there is always the defense factor. When someone says, roughly by this equation a .500 AL team would be a .570 team, are you using only defense stats from the days they play in the NL ballparks? Because you should, except for the extremely small sample size problem. What happens to Redsox defense when they slide Ortiz to 1B, Youk to 3B and Lowell to the outfield(or keep him out of the game). To take AL win % vs NL teams and extrapolate that to “this is what the team would do in the other league” is to ignore that teams are fundamentally built differently in each league.

In general, there are a lot of little things that probably make one large advantage for AL teams playing NL teams in inter league play versus if they played in the NL everyday. I haven’t looked at it yet, but what is the home-road splits for interleague play?

One final point, if this year the AL only wins 50% of interleague games this year, or some number smaller than the win percentage you used to come up with 22.5 runs, will you adjust the power rankings to use that percentage? Even then, I wouldn’t like the comparison that much, but at least it would be looking at this years competitiveness, not 5 years ago.

Sorry if I seem a little fired up about this issue, I really love all the stuff here on BtB, it’s just hard for me to believe a team like the Mariners would automatically win about 90 games if they were in, say, the NL Central. Ditto for the Angels with 5 outfielders for 3 spots, the Royals, the Twins, the White Sox, ect.