The Phillies Really Shouldn't Be Losing, And Yet They Are

The Phillies are in last place because of their terrible offense and their devastating injuries, right? Well, not exactly. Before the season began, most people thought the Phillies would win in 2012 the same way they won in 2011: with an okay offense that would score just enough runs to accompany their top-flight pitching. Entering last night's game, the Phillies' wRC+ (probably the most useful offensive statistic, since it's league and park-adjusted) was 94, seventh in the league. In 2011, the Phillies' wRC+ was 96, fifth in the league. So they've been a little bit worse, but only a little bit. That isn't enough to explain a dropoff from a 63% winning percentage to 48%.

Don't like the high-falutin' advanced stats? Then let's just look at runs scored. Last year, the Phillies scored 713 runs, seventh in the league. Entering last night's game, they had scored 236 runs, which, on a per-game basis, was eighth in the league. So again, there's been a bit of a dropoff -- a little bit more of a dropoff than that one-step change in the rankings indicates, actually -- but it still hasn't been very large.

The point isn't that the offense has been "good enough." The whole concept of "good enough" is meaningless, if you think about it -- everything's good enough and nothing's good enough. The Phillies' 2007 pitching staff was "good enough" for them to win the division despite being terrible, and the Phillies' 2006 offense was not "good enough" to get them into the playoffs despite being unstoppable. It's all relative. But we can say this much: this year's offense isn't significantly worse than last year's offense, and last year the team went 102-60.

So what has changed then? Well, it must be the pitching, right? But here's where the plot thickens. The 2012 Phillies' pitching staff is leading the league in xFIP and SIERA, just as the 2011 Phillies' staff did before it. Granted, it's leading the league by a somewhat smaller margin (in 2011, its xFIP was 0.14 better than the second-place team's and ~0.55 better than the league median, while in 2012, its xFIP has been "only" 0.09 better than the second-place team's and ~0.47 better than the median). But still, best is best. As with the hitting, the dropoff in pitching quality has been very small and doesn't do much to explain a 15-point drop in winning percentage.

Or so say the high-falutin' stats. Let's take a look at runs allowed and, oh... there's the problem.

In 2011, the Phillies led the league with a 3.02 ERA, #1 in the league, 0.19 better than the second-place team, and about 0.78 better than the median. Entering last night's game, their ERA was 3.81, #7 in the league, 0.80 worse than the first-place team (the Nationals), and only about 0.18 better than the median. That's an enormous dropoff, and it explains everything.

But how can that be if the DIPS have barely moved? Well, as we all know, discrepancies between DIPS and actual runs allowed are usually explained by weird BABIPs, LOB%s, or HR/FB rates. Here's a handy chart.

Wow, is that annoying.

Somebody else will have to crunch the numbers to see which statistic is most responsible for the huge dropoff in run prevention. But the one that really catches my eye is the HR/FB. 13.1% might not sound that bad (especially considering that HR/FB has been up across the league this year), but just as a temperature change of 1 or 2 degrees could have a massive ecological impact if spread out across the entire earth, a shift of a few percentage points in HR/FB is a huge difference when you're talking about the combined HR/FB of an entire pitching staff. The last time an NL team saw its pitching staff suffer from a rate over 13.0% over a full season was 2005, when it happened to both the Cubs (13.5%) and the Phillies (13.3%). In other words, it's a rate we haven't seen since the end of the steroid era. (And in case you were wondering, ESPN.com's park factors page says that Citizens Bank Park is once again not playing like a bandbox this year.)

So what does this all mean? That's sort of a difficult question. For one thing, HR/FB is sabermetrically a bit of an odd duck. They don't include it in FIP or WAR for a reason. It can be hard to wrap your head around the concept that a high or low HR/FB is purely the product of luck plus ballpark factors, and in fact, maybe it isn't. It sure seems like a high HR/FB is the result of mistakes or bad pitching generally. And yet, it is an observed reality that all pitchers tend to regress toward the league mean over the long term, and that even the few individual pitchers who appear to display a consistent ability to "beat" (or get beaten by) the mean only do so by a couple of percentage points at most. Whether the explanation for the Phillies' high HR/FB is that it's purely bad luck, or that it demonstrates that the pitchers have been making lots of mistakes, the fact remains that it probably won't continue. So we've got that going for us, which is nice.

And if it doesn't continue, the Phillies ought to be fine going forward. The Phillies' offense has actually outscored the first-place Nationals' offense by 0.3 runs per game this year, even though Nationals Park has been more run-friendly than Citizens Bank Park. If the Phillies' pitching staff can just prevent runs at a rate equal to what the Nationals' pitchers have done, then the Phillies should be able to win more games than the Nationals. And why shouldn't the staff be able to do that? It has a better xFIP and SIERA than the Nationals' staff does.

Of course, the problem is that even if we get the statistical regression we're hoping for, the Phillies are already in a six-game hole in the division. The Nationals are good. The Braves are good too. Even if the Phillies get better, you can't just spot six games to them and be confident about your ability to catch up.

Again, there are many lessons that can be drawn from all this, some wrong and some right. For instance, it would be wrong to "excuse" the offense, because, obviously, if the offense had scored more runs, the team would have been able to weather the dropoff in ERA. But it would be correct to say that the offense's struggles have "felt" like a bigger problem than would have been the case if not for what's happened to the pitching staff. The 2011 Phillies had almost as much trouble scoring runs -- you just didn't notice because they were winning in spite of it.

Likewise, it's correct, or at least defensible, to say that the team has played like crap this year. But it's wrong to say that the team's talent level has dropped off much. The underlying statistics do not support the proposition that the team's capability of winning degraded to any significant extent between 2011 and 2012.

Finally, it's wrong, and potentially dangerous, to say that the team must be blown up and overhauled. If this team's destiny is to win 81 games with the league's seventh-best offense and the league's best DIPS, then it isn't "really" an 81-win team. Now, this is not to say that the team cannot be a "seller" in any trade -- every individual transaction needs to be judged on its own merits, and it's theoretically possible for a specific proposed "sale" to make sense, especially when you consider that the team really is getting older. But if it's the type of sale that will set the team back massively in 2013 and 2014, then we shouldn't be in favor of that unless the long-run net gain is really large.