The Vuckovich Problem

Looking at the 1982 Cy Young Award Winner

The Vuckovich Problem

Pete Vuckovich had a great season in 1982.

Vuckovich went 18-6 for the Milwaukee Brewers, leading the club to an American League East title and a trip to the World Series. Vuckovich had an ERA of 3.34, struck out 105 batters, and showed remarkable stamina for a guy who was actually injured that September.

He’s also kind of become the stuff of tabletop legend.

Every once in a while you’ll see people talk about why games needs to create split ratings for pitching ability with runners on base. Usually they’re talking about pitchers like Vuckovich in 1982.

The classic example, of course, comes from Pursue The Pennant and Dynasty League Baseball creator Mike Cieslinski. I’ve spoken with Mike numerous times. He’s a great guy, and he’s one of the more active advertisers in the industry. He tends to say things like this about Vuckovich:

Cieslinski tends to focus on Vuckovich’s ability to keep runners from scoring once they reach base.

And, honestly, Vuckovich was kind of a unicorn in 1982. His 3.34 ERA came despite a batted against average of .275. The American League’s batting average as a whole in 1982 was .264, which means that Vuckovich was a little bit below average in giving up base hits. And, once you consider that Pete didn’t have to pitch against the Brewers, who hit .279 as a team, you start realizing that Vuckovich might not have been the dominant pitcher his ERA would indicate.

When you look at his splits, it seems that Pete was indeed much better with runners on base than with the bases empty:

I didn’t clean this table up very much, but you can see the difference, right? He had an OPS against of .801 with the bases empty, but only .671 with runners on base. He was a lot better at preventing home runs, at getting the strike out, at keeping his control, and at preventing hits.

It looks like a great argumet for some sort of “jam” rating, doesn’t it?

But there’s a problem.

The problem is that Vuckovich never actually demonstrated the ability to perform this feat in his career. In other words, this was likely a one off.

The real problem here is sample size. If we really believe that Pete Vuckovich had some sort of magical ability to pitch better with runners on base, we’d see a trend towards that in his career.

We actually see the opposite:

Once you increase the sample size and look at his ability over the longer term, you see that his skills in 1982 were likely an anomaly. For his career, Vuckovich was actually worse with runners on base than with the bases empty. In particular, he was a little bit worse at getting the strikeout, and had a tendency to give up more extra base hits.

Now, there are a few other things we can look at in 1982 to see if we can figure out what’s going on.

I made a post a couple of months ago about two advanced pitching stats: FIP, or Fielding Independent Pitching, and BABIP, or Batting Average on Balls In Play. You can check it out here:

The nice thing about these stats is that they help us provide some context to what Vuckovich did in 1982. In particular, they give us some clues as to whether he had unusually good fielding, or a healthy dose of what we call “pure luck.”

FIP is where we look only at pitching statistics that have nothing to do with fielding. We’re looking at the “three true outcomes” of strikeouts, walks, and home runs allowed.

Vuckovich struck out 105 batters in 223 2/3 innings in 1982, or about 4.2 batters per 9 innings. The American League average that year was 4.8 strikeouts per 9, which means that Pete was a little bit worse than average. In fact, the Brewers pitching staff as a whole struck out 4.4 per 9 innings.

Vuckovich walked 102 batters, or 4.1 per 9 innings. The American League average in 1982 was 3.2 walks per 9 innings, which means Pete was way off the pace.

Vuckovich gave up 14 home runs in 1982, or 0.6 per 9 innings. The league average was 0.9 per 9 innings, so he did a little bit better than average there.

All in all, when you do all the fancy math, Pete winds up with an FIP of 4.03. His ERA was 3.34, as I mentioned above. That tells us that Pete was either the recipient of excellent fielding, or of quite a bit of good luck.

We should also take a look at BABIP. Pitchers who have a lower BABIP than the league average are usually the recipients of good luck, or have benefitted from good fielding. Pitchers who have a much higher BABIP than the league average generally have had bad luck.

BABIP is interesting because it’s the opposite of FIP. In this case, we’re only looking at the batting average against for pitchers based on balls that were put into play. The theory here is that pitchers can’t really impact where the ball goes if it is in play. It’s kind of a measure of luck, and it actually works pretty well.

In general, if BABIP tells you that you were lucky or unlucky, there’s a good chance that you’ll see a reversion to the mean the next year. Or, in other words, if your BABIP was really low this year, it will probably be higher next year — and you’ll probably wind up looking like a worse pitcher as a result.

In 1982, Vuckovich had a BABIP of .299. The league average was .282. In other words, Vuckovich seems to have had poor luck on balls that were put into play. Maybe this means the Brew Crew didn’t field as well behind him as they could have. Or maybe it just means that he had an unusual tendency to give up base hits that season.

The Brewers pitching staff as a whole had a .284 BABIP, which is very close to league average. That tells me that their fielders weren’t performing magic tricks on the bases. Contrast that with the Ozzie Smith-led St. Louis Cardinals, who had a BABIP of .278 in the National league, where the league average BABIP was .286. Interestingly enough, the lowest BABIP in the National League in 1982 belonged to the San Diego Padres, who had a miniscule .259.

Vuckovich’s 1982 season is probably best symbolized by this game against Cleveland on July 29, where he gave up 10 hits in 11 innings but only 1 run. Vuckovich allowed runners to reach base all game long, and faced situations such as first and third in the top of the 8th and the bases loaded in the top of the 9th without giving up a run.

In the end, though, Vuckovich and the Brewers fell in the 12th:

Apologies for not finding a Milwaukee source; it’s one of the cities still underrepresented over at newspapers.com

Anyway, the long story short here is that Vuckovich does not appear to have had some sort of magical ability to not give up runs with runners on base in 1982.

In fact, if you look at the 1982 Brewers pitching staff as a whole, you’ll see that they were all remarkably good at pitching with runners on base:

That’s not quite as dramatic a difference as we see with Vuckovich’s 1982 stats, but it does at least indicate that something else is going on. Compare it with the 1982 American League in general:

It’s going to take some more digging, and it’s not entirely clear that the Brewers necessarily had some sort of magical fielding ability with runners on base. In fact, it seems to me that they just had enough good luck to squeak past the Orioles and finish in first place.

However, at the very least, I think we can clearly say that Vuckovich didn’t have some sort of magical clutch ability in 1982.

He pitched well, sure. In fact, he literally pitched his arm off that September. His career was cut short largely because of the way he was overused. It’s a remarkably tragic story.

However, Vuckovich was anything but a dominant pitcher in 1982. He didn’t have great control, he didn’t strike out a lot of guys, and he was more or less average in terms of home runs allowed. He also seems to have had a little bit worse luck than average in terms of base hits allowed, most of which seem to have come without any runners on base.

Honestly, I don’t think he was deserving of the Cy Young Award in 1982. However, I do think his 1978 season was likely deserving of the National League Cy Young Award, even though he didn’t get any consideration (mostly because the Cardinals were no good).

In the end, Vuckovich in 1982 is a great example of how small sample sizes and statistical noise can cause you to think you see a significant pattern that probably doesn’t actually exist.