All season, you've heard plenty - with much of it coming from here - about how good the Nationals' pitching has been. And it's true their starters have been consistent; the Nationals still haven't pulled a pitcher before the fifth inning, and 21 games into the season, they're the last team to do that.
So we know on most nights, the Nationals' starters have given the team a chance to win. But that's the minimum expectation. How good have the actually been? It depends how you look at it.
You can start by giving the starters this: They've been efficient. Livan Hernandez, John Lannan, Jordan Zimmermann, Jason Marquis and Tom Gorzelanny have combined to tie for the third-fewest pitches per plate appearance in the National League, with 3.78. By and large, this is a group of pitch-to-contact sinkerballers without strikeout stuff, and Zimmermann, the only one with a true wipeout pitch, hasn't been missing many bats this year; he's only struck out 13 in 24 1/3 innings.
(See what I did there? I started that paragraph with an advanced stat - it's true; it's in ESPN's Expanded II pitching stats category - and then finished it with a sentence cramming as much hackneyed scout-speak in as I could. Yep, I'm an equal-opportunity baseball wonk.)
Anyway, we know the Nationals starters have done an effective job of pitching economically; on most nights, they've used their fastballs to pitch ahead in counts. And if there's one thing they deserve praise for, it's holding down walks - the starters have walked just 36 batters at this point, second fewest in the National League only to the Phillies' vaunted staff.
But as efficient as the group has been, it's yet to produce the kind of dominant outing that's good enough to win a game without much help from the offense.
BIll James' Game Score stat measures each starter's outing by awarding a set of points, so we can compare starts from one pitcher or year to another (we're back in sabermetric-land for a minute; I'll be back for the old-fashioned types in a minute). Under his formula, each pitcher begins with 50 points - so we can call 50 an average game score - and gets additional points for innings and strikeouts. The pitcher loses points for runs, hits, homers and walks.
Now, there's a bit of bias in the stat in that it puts a high weight on strikeouts, and if you know anything about the Nationals' pitching staff, it's that it doesn't exactly jive with the miss-bats approach of most sabermetricians. Still, the Nationals' starters don't score particularly well here; the only two Washington starters who have average game scores in the NL's top 40 are Marquis and Zimmermann. They're at 51.5 and 50.8, respectively, so in other words, they've done a consistent job of being above average, but not dominant.
Let's apply the eye test here: there haven't been many nights where you've been able to leave the park saying you just saw the kind of pitching performance you'll never forget, unless it was by Cliff Lee or Roy Halladay against the Nationals. That's not to say that Washington's starters have to be ethereal, or that they're even going to be; this group won't have that kind of starter until Stephen Strasburg returns, and depending how healthy he is, he might not even be that pitcher.
But as well as the pitch-to-contact approach has served the Nationals so far this year, it's susceptible to flaws from a defense that's still making far too many errors (18, the second most in the NL). And it means that on nights where pitchers don't have their best stuff, they mostly lack the strikeout ability to get out of a jam. That's why manager Jim Riggleman calls on Tyler Clippard so often in tough situations; he's the one Nationals pitcher that misses bats so consistently, you can tell him, "We need a strikeout," and know he's got a reasonable chance to deliver one.
As the weather gets warmer and hitters get hotter, the Nationals' starters could fall back to the pack. We've even seen that in the last few days, as the Nationals have lost four of five; their only quality start in that time came from Marquis, who allowed three runs in six innings on Sunday and fought his stuff all day to even get to the bare minimum for a quality start. The Nationals' starters have done a good job of generating weak contact (opponents' batting average on balls in play is only .285, tied for fifth best in the NL), but that may not last.
In the end, though, no one expected much from this group. It's got two veteran starters (Hernandez and Marquis) who would be at the back of a rotation on a good team, and though they're younger, Lannan and Gorzelanny are probably in that realm, too. Zimmermann is the one who has a chance to be special, and we saw some of that when he took a perfect game into the sixth inning against the Phillies. He'll only pitch 160 innings this year in his first season back from Tommy John surgery, though, and though he's been efficient working under a tighter pitch limit, he could stand to produce a few more strikeouts.
But as a whole, an area of the Nationals' roster that most worried could be a problem has instead been reliable - not great, but reliable. The degree to which the starters have exceeded expectations is probably what's producing most of the praise. And if that continues, it's about as much as the Nationals can ask for.