Pitching a Fit: AL Style

Dave Paisley

After taking a look at pitching trends in the National League last time out, here's a look at what happened in the AL last year with a view to what's likely to happen this coming season.

If you cast your mind (or browser) back to Friday's article, you'll remember that I took a look at how a pitcher's ERA correlates with the OPS he gives up. The correlation is extremely strong, and it isn't terribly surprising. Just as a hitter creates runs by banging out hits and taking walks, some pitcher is surrendering those hits and walks. I like to think of OPS (On base percentage Plus Slugging average) as the building blocks of runs.

If two pitchers give up the same OPS, the only reason they would give up different ERAs is if there was some difference in luck from "average". For instance, a guy that gives up three hits in a row then gets nine guys out is going to give up a couple of runs. The guy who gives up a hit then gets two guys out repeatedly is unlikely to give up any runs. Does it mean the first guy is a worse pitcher? Not really. It just means that he's unluckier. Over the course of a career that luck will tend to even out. Usually it will even out even in a single season, but not always.

I classify single season luck as being a significant deviation from the predicted ERA predicted by OPS given up. Take a look at the following chart.

Taking pitchers who have been around for all or most of the last five years who pitched in the AL last year, I used the data from 113 pitchers to predict an ERA for last season based on OPS allowed. Some pitchers allowed more (much more) than the prediction, while others allowed less. Those deviating significantly from the prediction I classified as lucky or unlucky, depending on the direction of the deviation. The red line shows the line of neutral luckiness. Those on the high side were unlucky, those on the low side were lucky.

While Mark Clark had a tough season with injury and general pitching suckiness, he was also, if you can believe it, unlucky. Based on the OPS he coughed up, he should have only surrendered an ERA of 7.00 or so. Instead, he was up in the 8.50 range. Sure, it was a small sample, but can you believe he could have that bad a year again? Well, sure he could - his arm could fall off.

Another unluckiness candidate was Joey Hamilton, also a small sample because of injury. It's interesting that injured guys tend to also be unlucky, so maybe there's something in that. Other limited action unlucky pitchers were Jaret Wright and Jimmy Haynes.

I might be tempted to say that Minnesota pitchers tended to be unlucky, but LaTroy Hawkins and Eric Milton's bad luck is balanced by Brad Radke's good fortune.

Whatever the reason, it seems that the above mentioned pitchers are almost bound to have better luck in 2000, even if they pitch pretty much the same as last year.

On the other hand, beware the lucky pitchers. Radke, Allen Watson, Mike Mussina, Wilson Alvarez, Rick Helling and Bret Saberhagen all allowed substantially smaller ERA than their numbers warranted, so don't expect that to hold.

That said, let's take a look at how AL pitchers changed significantly from their long term numbers. The following chart shows all AL starters who've been around a while and who pitched last year. It highlights by name the guys who had significant differences in ERA from their five year average. The red line is the mean through all of the data, which shows that, on average, ERAs were up almost ten per cent in 1999 over the average of the last five years.

Right at the top is our friend Mark Clark. With a five year average of 4.50 or so, he managed to top that by almost doubling it last year. As we saw from the previous chart, more than half of that change was due to just getting hammered, while the rest was just bad luck. Whatever it was, do you think he'll bounce back? Like a sack of wet cement on a sidewalk, I'm guessing.

Then there's Jeff Fassero, whose 1999 performance was about as successful as jumping out of a plane without a parachute. Prospects for a rebound are better than Clark's, but age could be a factor here. Red Sox fans - don't hold your breath. If Fassero could have turned in an ERA under 5.00 he would have been looking at an almost Chuck Finley-like contract, but instead he gets $2M. Oh well.

Other big cliff divers include Tim Belcher, Roger Clemens, Kevin Appier, Mark Portugal, Mike Morgan, Dwight Gooden and Jaime Navarro. Hmm, I'm starting to smell the stench of washed up veterans here. Not that some of them won't be useful, but these aren't guys you want to bet the farm on in a couple of years time. We'll give the unlucky Joey Hamilton the benefit of the injury doubt, while Justin Thompson may have just been having a bad year.

On the plus side, we see lucky boys Brad Radke, Bret Saberhagen and Allen Watson, so we can discount actual performance there to a degree. That leaves only Jeff Suppan and Pedro Martinez who beat their last five years significantly. (Suppan hasn't been around that long, of course.)The chart just makes Pedro's 1999 look even more spectacular.

about the author

Dave Paisley iwould like you to know he'd still be a major league pitcher today, if it weren't for that microchip the Russians planted in his left shoulder. Ask if the crow flies at midningt at drdjp@strikethree.com.