The importance of relief pitching, cont'd
 
In my recent article about the consistency and importance of bullpens, I pointed out that there's not much difference between the playoff teams in terms of their league rankings in offense, starting pitching, and relief pitching. The implication was that relief pitching may have become as important as the other two elements of the game even though pitching is only half the game and relievers account for only one-third of all innings.

Even since that article was posted, I've been thinking that more needed to be said about other possible interpretations of the data, and several astute readers have written with pertinent comments in that regard.

The common thread in my thoughts and those reader comments is the need to refrain from assuming that correlation means causation. In this case, high bullpen rankings and postseason play were highly correlated, but we need to be careful about claiming that teams made the playoffs and had success in the playoffs because their bullpens ranked very high.

Let's consider some other possibilities.

First, it's a very small sample, only five years, so these results may not be typical.

Second, my analysis was based on simple rankings. If, however, the spread in OPS between the top and bottom teams was much greater for bullpens than for offenses, we're not comparing apples to apples. A first-place ranking in bullpen OPS might represent a more significant advantage over the league than a first-place ranking in offensive OPS.

Though I didn't write about it, I did look at these spreads before publishing the article. They were the same for starting pitching and relief pitching, and only a little smaller for offenses, so I don't think this was a major factor.

Third, it's possible that some teams are successful in more than one area of the game. Perhaps they have very good starting pitchers and can spare some quality arms for bullpen work. Perhaps they have the financial resources to buy talent of all types or the management skill to develop talent in all three areas. If there's a strong link between high-ranking bullpens and high-ranking offenses/rotations, it might be the offenses/rotations that are keying their success, with the bullpens just along for the ride.

Let's take a look at a few examples to see if we can shed any light on this third possibility, focusing on the AL, mainly because I don't have time right now to do both leagues. (As was the case with the original article, all rankings are based on OPS.)

In 1998, the Yankees were first in offense, first in starting pitching, and third in relief pitching. They notched 114 regular season wins and won the World Series, losing only two postseason games along the way. Their only serious competitor was the Red Sox, who were fourth in offense, third in starting pitching, and tops in relief OPS. The Boston bullpen advantage didn't amount to much, however, as they finished 22 games back of New York and lost the division series to Cleveland.

The next year was similar. New York was third in scoring, second in starting pitching, and first in relief pitching. This time, they lost only one postseason game, and their strong bullpen may have been the difference this time. Cleveland was first in offense and fourth in starting pitching, so the Yankees didn't have a significant edge in these areas. But the Indians were only seventh in relief pitching, and that bullpen was a major reason why the Indians lost the division series to Boston.

In 2000, the Yankees were second in the league in relief pitching, and that was enough to carry them to another World Series championship, even though they were only sixth in offense and fourth in starting pitching. The White Sox took the Central from Cleveland even though the Indians topped them in both offense and starting pitching. Two things put Chicago over the top -- a very efficient offense that outscored the Indians by 28 runs despite the deficit in OPS, and a stronger bullpen. In the West, Oakland and Seattle made the postseason despite finishing tenth and sixth in relief OPS, respectively. Overall, this was a mixed bag in terms of the importance of relief pitching.

In 2001, the Yankees were again carried into the World Series by their bullpen. (Ironically, they lost game seven when Mariano Rivera couldn't hold a ninth-inning lead.) The New York offense was only eighth and the starting pitchers only fourth. The Red Sox were better in both categories, but their bullpen fell to ninth in the league, and the team barely managed to top the .500 mark.

In the Central that year, Cleveland rebounded to win the division despite their weak starting rotation, which was tenth in the AL in OPS. The offense was second-best, however, and that, combined with a fourth-ranked bullpen, was enough to give them the division title. If you combine the offensive and starting pitching rankings, Chicago and Minnesota were right with Cleveland, but they couldn't match the Indians pen.

The West was clear cut. Both Seattle, which won 116 games, and Oakland, which also topped the century mark, were well-rounded teams that finished among the leaders in all three areas. Seattle didn't just lead the league in bullpen OPS, though. It dominated, posting a .611 figure that was by far the best since 1998 in either league, and that was a huge part of their success.

In 2002, all four playoff teams were in the league's top seven and Anaheim's league-leading pen was instrumental in their World Series run. On the other hand, Seattle was second in relief OPS and Chicago fourth, and both teams were on the outside looking in come October. Minnesota and Oakland were still playing despite finishing sixth and seventh, respectively, in bullpen OPS.

In the fifteen AL postseason series that have been played since 1998, the team with the higher-ranking offense (based on regular-season OPS) won only five times. The team with the higher-ranking starting pitchers won seven times. The team with the higher-ranking bullpen won twelve times. In four of those twelve series, the winning team was victorious even though their opponent was higher-ranked in both offense and starting pitching.

By reviewing these five years, I don't claim to have done a comprehensive analysis of the relative importance of offense, starting pitching, and relief pitching in today's game. These examples provide some evidence to support the idea that relief pitching is right up there in importance, but they provide some evidence to the contrary as well. I will say that relief pitching is more important than I thought before I began looking at these rankings.

If I was responsible for building a team, my priorities would be starting pitching, then offense, then relief pitching. That was true a month ago and it's true today. The difference is that I previously viewed relief pitching as a distant third on that list, with starting pitching only a hair more important than offense. Now I'd say that relief pitching is a very close third.

(By the way, I don't mean to ignore defense. It's a very important part of run prevention that affects the rankings for starting pitchers and bullpens. One cannot build an offense without thinking about the defense at the same time.)

posted by Tom at 11:50 PM