Don’t Blame Pitching — D-backs Defense Either Much Worse, or Extremely Unlucky

More than a third of the season is in the books now, and the D-backs have played a majors-leading 59 games. The team’s offense has been at least decent, but despite better pitching numbers in May, the D-backs have allowed 4.95 runs per game, easily the worst in MLB. The pitching staff has definitely struggled, but might there be more to it than that? There’s evidence to suggest that the D-backs defense has been much less efficient at turning batted balls into outs, with very large implications.

The D-backs actually rank dead last in the majors in ERA (4.47), but 26th in FIP (4.07). That’s a smoking gun that team defense has not been an asset so far this season — FIP stabilizes quickly on a team basis, and only 6 teams have a greater discrepancy between ERA and FIP. It also looks bad if you don’t account for errors. In runs per nine (RA9), the D-backs also rank dead last (5.00). And if you compare RA9 to FIP, the D-backs have a discrepancy of 0.94 — second largest in the majors. The D-backs have been spectacularly inefficient at turning batted balls into outs this season.

A very large part of the blame for the team’s ERA or RA9 is home runs — the D-backs have the second-worst HR/FB ratio in the majors (13.5), and that alone has made the D-backs pitching staff appear worse than expected. The team ranks a surprising 11th in xFIP, which is essentially the team’s expected ERA if batted balls were converted into outs as expected (like FIP), and if fly balls turned to home runs at a league average rate (unlike FIP). The D-backs ranked 13th in xFIP in 2013, and so it’s fair to say that taking defense and HR/FB out of the equation, the D-backs pitching staff has been better this season than it was last season.

And yet, this season the team ranks dead last in ERA despite ranking 17th last season. It stands to reason, then, that the only things making this season’s pitching staff worse are defense and home run ratio.

The reason we frequently use both FIP and xFIP is that we don’t really know how much control a pitcher or pitching staff has over the conversion of fly balls into home runs. Fly balls are mostly about launch angle off the bat (vertical), but whether the ball is “barreled” and hits the sweet spot (horizontal) is also very important. We know pitchers can control launch angle to some degree (pun intended), because ground ball rates (and fly ball rates) tend to be sustainable (just ask Brad Ziegler). It stands to reason that they could also control sweet spot rates (paging Mariano Rivera and his bat-breaking cutter), but we aren’t able to measure that as well right now. The data from Hitf/x technology hasn’t been released to the public like the data from Pitchf/x.

The line between a “line drive” and ground and fly balls is primarily, but not exclusively, about launch angle. There’s a bit of “hard hit” in there, as well. And we know from experience that, as they’re currently classified, about 70% of line drives are hits, as well as 20% of ground balls and just 12% of fly balls that stay in play. So let’s look to batted ball rates.

What’s this? The D-backs have had decent fly ball rates, but a higher percentage of their fly balls have turned into home runs. Ok — we know that already. As for what the hell is going on there, that’s something I intend to look into, but not now. Today, we’re primarily interested in what the hell is going on with the defense, and because there’s a discrepancy between ERA (or RA9) and FIP (forgetting xFIP for a now), we know that home runs aren’t the only story. The defense has also been inefficient.

But look again at these numbers. Fly ball rate has stayed almost exactly the same, but the pitching staff has traded a bunch of line drives (of which about 70% tend to fall for hits) for ground balls (of which about 20% tend to fall for hits). Batted ball numbers suggest that BABIP should have gone down this year, but BABIP actually went up — by a lot. This is evidence that the defense has been less efficient at turning balls into outs.

Indeed, defensive metrics are way down on the Diamondbacks. Compare last season with how the team has fared so far:

Defensive Runs Saved and Ultimate Zone Rating both thought that the D-backs were excellent last season. Not shown is that in both metrics, there was a pretty steep difference between second and third. They weren’t the best overall, but they were still an outlier, in a good way. Defensive Efficiency (from Baseball Prospectus) kind of begs the question, since that’s what we’re trying to determine — but it tells almost the exact same story as BABIP. In BABIP, the D-backs ranked 13th in 2013 (.292), but are 25th this season (.307).

Defense this season has stayed steady in some components, like the runs-saving effects of outfield arms (at which the D-backs are still great). The principal shortfall has been in “good fielding plays” and defensive range in general, but neither of those have rated especially poorly. So it’s not that defense has been terrible, exactly, but more that in addition to not being excellent like last season, it’s no longer a positive. The D-backs defense is now mediocre, at least based on a third-season’s worth of data. It’s also been unlucky, as the defensive metrics DRS and UZR are essentially neutral on the D-backs, but in defensive efficiency, the team has been quite bad.

One last check — we have Inside Edge fielding data available at FanGraphs now, so let’s see if this shortcoming is just in hard or easy plays, etc. Inside Edge stats are broken down by the difficulty of the play attempted — in the column on the far right we see the D-backs’ efficiency with batted balls that players turn into outs between 90% and 100% of the time.

The D-backs are a little down across the board so far this season — and it would be hard to match their excellence with particularly difficult (1-10%) and fairly difficult (10-40%) batted balls. But the D-backs have gone from pretty good to amazingly horrible in turning fairly easy batted balls into outs.

Do not underestimate the impact of this shortcoming. The D-backs have attempted plays on 221 balls that Inside Edge thought there was a 60%-90% chance of making. The difference between last year and this year (14.5%) is 32 balls that were hits (or errors). Let’s pretend all 32 of those are singles/first base errors, which is the most conservative way of looking at it — in that case, the D-backs’ failures with fairly easy plays have resulted in about 15 extra runs, or nearly two extra losses.

And it’s not just fairly easy plays that we’re talking about. Overall, A.J. Pollock and Chris Owings have been tremendous defensive assets this season. Ender Inciarte and Martin Prado have been very good as well, especially when left at their best respective positions. Mark Trumbo was worse than expected, and Aaron Hill has also rated poorly this season despite some excellent work (see 7th bullet here). But Gerardo Parra, Cody Ross and Paul Goldschmidt, like the team in general, have gone from being defensive standouts to merely mediocre, at least in this small sample of 59 games. The team’s defensive play is definitely something to watch going forward — and although the club has done extremely poorly in run prevention this year, team defense seems to have played at least as significant a role as pitching.