Or is it something else, like the way they’re combined?

Offense: a touch below average (.327 wOBA)
Fielding: a touch below average (-9 runs)
Pitching: solidly above average (4.11 tERA)

(All pre-league adjustment, natch.)

Wow, this is a pretty angry conversation about something that I’ve been doing most of the season. I will say that I have noted that the strongest critiques are almost always from fans of teams ranked poorly. FWIW, my Reds come in last, while my most-hated team, the Yankees come in first. ::sigh::

Anyway, some points:

1. The league adjustments are based on interleague records, and (loosely) the studies done on players swapping leagues. The two match up. And those league-change studies are the basis of the typical adjustment of a half-win a season for AL players vs. NL players most people use (though FanGraphs does nothing at this point, afaik).

2. If you don’t like the league adjustments, just look a the eW%lg column. You can click on it to sort by it. That column essentially assumes no difference in quality between leagues. I don’t think that’s true, but it makes some people happier.

3. If you’re going to compare actual records to expected records, you really should use the eW%lg column. It’s quite literally a component pythagenpat estimate of W%. The eW% column assumes league adjustments, which is much more hypothetical (what if you took all the AL and NL teams and threw them into one league).

4. Sorry to hear some folks don’t like the name of the list. This is the first time in 3 months someone has objected to it, so I think I’m going to stay with it.

5. As Sky mentioned at BtB, there are a lot of things not in the calculations yet: baserunning (beyond SB & CS), bullpen leverage, and strength of schedule being the biggest of them. And there are reasonable criticisms of how I’m implementing tRA, park effects, etc. Happy to talk about that. My feeling is that most of those things, if implemented, will have pretty small effects. But I don’t know that.

6. As for the Cardinals, through Tuesday, here’s what I have:
True W%: 0.562
Pythag W%: 0.548
eW% lg (without league adjustment): 0.504

So yes, definitely a “miss” low. In fact, it’s the second-largest “miss” in baseball, behind only the Giants. Why? Let’s look at runs scored and runs allowed:

Actual RS (after park adjustments): 556
Estimated RS (after park adjustments): 542

So eRS is 14 runs shy of the actual total. I don’t know why that is—I literally just take actual RS and wRC from fangraphs and park-correct both of them in the same way. Could be lucky, could be baserunning, could be something else. That’s not a small disparity, but it’s not huge either at this point in the season. We can add in some baserunning, but BPro has them (looking it up) at +1 run above average, so that doesn’t help much.

Actual Runs allowed (park adjusted): 502
Estimated RS (park adjusted): 537

So there’s the biggie. I’m estimating 35 more runs allowed than actually has happened. Looking quickly, the Cards tERA is 4.11 whereas their FIP (as I do it, anyway) is 4.04. Pretty close match there, but if I used FIP instead of tRA to estimated runs allowed it comes in at 523 RA. Definitely better.

Fielding-wise, UZR at fangraphs has them (or did through Tuesday) at -13 runs. THT’s team fielding stat, which is a mesh of DER and batted ball data, has them at -3 runs. I’m using the average of the two, which is -8 runs, in the power rankings. You could argue we should drop down -3 runs if you don’t like UZR, though I think UZR’s a better fielding estimate than THT’s (and studes has concurred with me on that point in the past). Thinking about it, though, I’m not including Molina’s defense in any of this, so I think that would help justify going to the -3 runs.

So if we do everything in a direction that makes the Cards look better (which is not exactly the objective way to make a decision, but whatever), we get an estimate of 518 runs allowed vs. 543 estimated runs scored. That would put the component pythag at 0.524. Closer to reality than the 0.504 I had before, but still shy of the actual W%. And again, I’m doing everything here to make the Cards look better, which almost certainly will have the opposite effect on other teams. In other words, these decisions aren’t necessarily the “correct” ones just because the make the Cardinals look better.

So, I guess what I’m saying is that why I can certainly sympathize with folks being upset that your team is ranked lower than you’d like, I think it is the case that, at least based on the component stats I’m using, that the team might be “overperforming.” Whether that’s due to lucky, leverage, or something else, I have no idea. But that’s what the BtB power rankings do—they give you something else to look at when evaluating the relative strength of teams.

Hope that helps. And look on the bright side—with all of your additions this season, the Cardinals are almost certainly a better team than their component stats would indicate! :) Enjoy the ride.
-j