You Complete Me

Is it just me, or is the ghost of Steve Carlton posessing most of the Florida Marlins' starting rotation?

Actually, Lefty's still alive, so it can't be his ghost. For that matter, it's not just the Fish either, it's most of the National League. Looking at the box scores from the last week or so, it seemed to me that there was a disproportionately high number of really impressive performances by starting pitchers. It seemed like every day there was another complete game, another shutout, aother 8-inning, 2-hit, 9 strikeout box score, and it didn't seem right.

Or, perhaps more accurately, it didn't seem normal. Isn't this supposed to be the Age of the Hitter? Isn't this supposedly the greatest offensive era in history? A time when a 40-year old with bad knees can win a batting title and get walked out of sheer fear more often than anyone else in his league gets a hit? An era in which everybody (it seems) can hit 20 homers in a season? When even a 5'9", 178-pound secondbaseman can slug almost .900, at least for two weeks? Where's the offense, man?

So, to figure out what was going on (I know this is gonna be hard to believe, but stick with me here...) I did some research. I looked at the contributions of starting pitchers through the first two weeks of this season and compared the numbers to last year's overall numbers. (Just comparing to the first two weeks of 2004 might have been more appropriate, but hey, what do you expect from someone who does this in his spare time?) And here's what I found:

I'm wrong.

Well, I'm wrong about the American League, at least. Take a look:

  G      IP     IP/GS   ERA  CG  CGF  SHO
 170   1000.33  5.88   4.39   5   34   0
2266  13352.33  5.89   4.83  79   29  33

American League starters have averaged almost exactly the same number of innings per start (IP/GS) this season as they did all of 2004. Though there has not yet been a complete game shutout pitched in the Junior Circuit in 2005, they only happened about once every 69 games last year, so if two of them happened tomorrow we'd be nearly right back on pace. The frequency with which complete games occur (CGF) is actually down just slightly from 2004, as we've had a CG only once every 34 games this season, compared to once every 29 last year.

The one interesting thing to note here is that run scoring, at least scoring of earned runs against starting pitchers, is significantly down. Last year's AL starters had an ERA of 4.83, and so far in 2005, AL hitters are crossing the plate at a rate almost half a run lower than that. Unless there are just a lot more errors being made, this seems odd. I guess it seems odd either way.

The National League, however, may be a slightly different story:

  G      IP     IP/GS  ERA  CG   G/CG  SHO  G/SHO
 184   1105.34  6.01  4.05   9  20.44   7   26.29
2266  13236.66  5.84  5.05  71  31.92  36   62.94

The Senior Circuit is showing a slight increase in average innings pitched by starters, from 5.84 last year to just over 6 this year, or a difference of about half an out per game. Starters' ERA is down an entire run, to 4.05, a huge drop from last season. Furthermore, complete games are happening much more often, every 20 games or so, compared to every 36 games last year, and complete game shutouts have occurred more than twice as often, ever 26 games compared to every 63 last season. Suddenly, at least in the National League, starters know how to pitch again!

Why might this be?

The first thing that occurred to me was: Barry Bonds.

Barry hasn't played, and he is an incredible offensive (and sometimes, incredibly offensive) player, so maybe he's the difference, right? Nope. Even if Barry is worth two runs per game more than whatever stiff they're trotting out to left field in his stead (he's not), that would still only bring the NL ERA up to 4.23, still more than 3/4 of a run below last season.

On a broader scale, maybe it's the steroid testing policy? I mean, if the players were afraid of getting caught and stopped taking them, then naturally you'd expect offense to be down, right? We can't know just how many players aren't taking steroids anymore, so there's really no way to measure this directly. On the other hand, (there's a wedding band,) and it's not as though the steroid policy has turned up anyone significant yet. The only two players to serve suspensions (Tampa Bay CF Alex Sanchez and Rockies CF Jorge Piedra) are unimpressive players on unimpressive teams, no one who might have been expected to alter their whole league's run-scoring abilities. (In his defense, as unimpressive as the Devil Rays may be, they have a better record than my favorite team...but then so does everyone else in the AL.)

Maybe the anomaly isn't this year, but last year. The rate at which complete games are occurring in the NL works out to about 127 over the course of the season, and even though the last year the NL saw that many complete games was 2000, last year's rate of 71 was an all-time low, by a significant margin, as the second-lowest occurred in 2001, with 96. Similarly, the American League total of 79 in 2004 was also an all-time low, way below the previous record of 103, also set in 2001. Without needing to pinch-hit for the pitcher late in a game, you'd expect the AL to have more complete games in general, even though run scoring tends to be lower in the NL.

So maybe all we're really seeing is the return of some sanity to major league baseball, in National League parks, if not in the American. Pitchers are supposed to be able to finish a start once in a while. It should be possible for a starter to pace himself and not have to exert all his available energy with every pitch, thereby ensuring that he will never ever be able to throw more than a hundred pitchers, and certainly never finish a game. If it really is the steroid policy, and some of the edge hitters have enjoyed for the last decade is beginning to recede, then I'm grateful. If pitchers can get batters out and complete a start without throwing 120-140 pitches to do it, the I want to see more of it. It will be better for everone involved.

Of course, two weeks into the season is not the best time to do a retrospective on it, so we'll need to check back in this October, and see if the policy is really helping. If NL starters are still holding a collective ERA around 4.00 and finishing a start every 20 games or so, I'll be happy even if the only other players suspended are more Darryl Hamilton than Darryl Strawberry.