Girardi versus Torre in bullpen use

There is a common narrative when it comes to looking back on the two great Yankee managers over the last eighteen years. The narrative goes something like this: Joe Torre burned out his bullpens and Joe Girardi's use of his bullpen is one of his strengths as a manager. I have heard variations of those memes over the years and wondered if there was a way to measure the bullpen usage to see how true these narratives were. Once I put together all the numbers, the conclusion I came up with is that Torre really did not earn his reputation until the last five years of his tenure. First, let's look at the most basic of numbers, things like ERA, WHIP and bullpen losses. Each provides some insight to the conclusion.

Joe Girardi's bullpens have beaten the average American League bullpen in ERA every season he has been the manager. His accumulative average bullpen is then higher than the league over that time period. Joe Torre's bullpens beat the league average for his first eight seasons in pinstripes. In 1997, the bullpen ERA was almost a full run below the league average.

But in Torre's last five seasons, four of those five bullpens actually finished above the league average in ERA and the one season his came below (2006) was only by three lonely points. Here are their accumulated averages:

Manager - ERA - Lg ERA - Diff

  • Torre - 4.00, 4.33, -.33
  • Girardi - 3.56, 3.88, -.32

As you can see, the two numbers are nearly identical. Torre would have been much higher if the last five years were not so tough. It should be noted that Girardi's last three years were not as effective as this first three. The lower Girardi average ERA is in a much more pitching environment than the offensive days Torre managed, so don't make a big deal of that difference.

The same thing shows up in the accumulated WHIP averages. Torre's bullpen WHIP was much lower than league average for his first eight years and then was actually higher in two of his last five years. Girardi has again struggled a bit the last three years with his bullpen WHIP but the numbers come out nearly identical:

Manager - WHIP - Lg WHIP - Diff

  • Torre - 1.359, 1.424, -0.065
  • Girardi - 1.271, 1.337, -0.066

I'd call that a wash. So far, we have seen little difference between the overall results of the two managers and their bullpens. It should be noted that we are judging all relievers equal here and not taking into account the kind of talent each manager had. But it is a manager's job to make the most of their bullpens and these numbers show that both have done about the same in these categories.

Now let's look at two other basic categories, the number of losses the bullpens sustained per year under each manager and the percentage that inherited runners by relievers scored:

Manager - Losses per season - Inherited runners scoring

Torre - 17.67, 33%

Girardi - 17.5,  27.7%

While the losses per season are similar, the inherited runners scoring favors Girardi. I did not do league averages for these stats though and the higher scoring environment in Torre's years may play a factor.

What about how the bullpens were used? I then looked at things like days pitched without rest, number of outs per outing and times that more than three outs were asked of the reliever:

Manager - Days without rest - ave number of outs - more than three outs

  • Torre - 93.33, 3.52, 135
  • Girardi - 89.83, 3.20, 115

Some notes on the numbers above: Torre's early years saw similar days without rest numbers as Girardi. The difference really came about in the last four of Torre's seasons. Three of those four seasons showed the number of days pitched without rest jump to over a hundred and the fourth was at 92. 2004 was really bad as it happened 131 times! In Girardi's six years, he had one season (2012) where that figure jumped to 117, the second highest of the eighteen year period. But all other years including 2013 were in the eighties and seventies.

Torre left his pitchers in longer. The number of outs per outing and the times when more than three outs were required showed a slightly different philosophy. All three of these numbers in this last sequence seem to favor the idea that Torre expected more out of his bullpen and Girardi is more conscious of rest and length of outing. But again, Torre seemed to demand more with less rest the further in the rear view mirror the 2000 championship season was toward the end of his tenure.

The one constant for both managers was Mariano Rivera, except for 2012, when Rivera missed most of the season for Girardi. Joe Torre should be given credit for his emphasis on trying to make each game a six or seven inning game where the bullpen shut it down from the seventh inning on. Girardi has tried to continue that tradition. That emphasis really worked for Torre in his first four seasons with Rivera, John Wetteland, Jeff Nelson, Mike Stanton and Ramiro Mendoza all factors in those first four years.

But after Nelson and Stanton broke down, Torre simply did not have the same success in his later years and struggled to find a combination to put in front of Rivera. Girardi used the tandem of Rivera and David Robertson to great effect as long as they were healthy and often had a decent option for the seventh inning. Injury problems led Girardi to improvise and 2012 was a real scramble act in particular.

The conclusion would seem to be after looking at everything that Joe Torre earned his bullpen wrecking reputation in the last four years of his tenure. Up until that point, his usage of the bullpen was similar to Girardi's and the overall numbers seem to show that Torre was not as bad on his bullpen as we thought.

The bottom line is that both garnered similar results in bullpens that regularly beat league averages in ERA and WHIP. Perhaps Torre should be given a little more credit than he gets with his bullpen usage and perhaps Girardi should get a little less.