6 responses

1. 

   Tom MacDonald July 28, 2013 at 5:41 pm | Reply

   Perfect visual description of the Tigers’ bullpen – Smyly and Benoit Top 5, no one else in the Top 150. Little wonder Leyland sounds crabby when speaking about his bullpen.

2. 

   AD July 29, 2013 at 8:39 am | Reply

   1. Reinforces what other metrics indicate about Papelbon: http://aldland.wordpress.com/2013/06/25/bay-of-cigs-forget-what-you-know/.

   2. Why do you think Kimbrel doesn’t rank higher?

   3. Not that they’d trade him to a division rival, but do you agree the Tigers should pursue Crain, as some have suggested?

   1. 

      Neil Weinberg July 29, 2013 at 10:32 am | Reply

      Papelbon has tremendous numbers in his career, but in his current form his isn’t enough of an upgrade given the cost. Crain would be a good addition if not for the injury. I woulnd’t risk much because he isn’t going to pitch for at least another week and it’s unclear when he’ll be back. The Tigers just got Veras, so that’s probably it.

      As for Kimbrel, he has only inherited 1 baserunner (whom he allowed to score). This metric does not penalize Kimbrel for that, but it also doesn’t give him credit for stranding runners. He ranks very well in the other components of the metric, but is given a league average score in in Strand Rate. Kimbrel simply isn’t used with runners already on base. If he were to have inherited more than 5 runners and stranded more thna 85% of them, he would be right near the top.

      1. 

         AD July 29, 2013 at 12:02 pm

         That makes sense re: Kimbrel. I’m still relatively new to advanced metrics, to say nothing of SOEFA, and the calculations behind them. Are you concerned that SOEFA may undervalue guys like Kimbrel, who are only ever going to come in at the beginning of an inning? Do you think there’s a way to build in an element that rewards guys who a) are extremely unlikely to inherit any runners due to the manner in which they are used and b) also do not usually allow runners to reach in the first place?

   2. 

      Neil Weinberg July 29, 2013 at 12:56 pm | Reply

      First of all, if you haven’t seen it, this is the full explanation of SOEFA and how I calculate it. https://newenglishd.com/2013/06/26/a-new-way-to-measure-relief-pitchers-soefa/

      That will probably answer any questions but here’s the jist. SOEFA combines three aspects of a reliever’s job:

      Stranding runners
      Not allowing runners
      Not giving up runs.

      Those are pretty basic functions of the reliever and I’m trying to create a stat that combines some context specific info (when they are used, how well they handle big spots) with some context neutral info (how often they K guys, not just when).

      For each of those three qualities, I give each pitcher a score based on how much better or worse they are from league average. Therefore, a pitcher with a great strand rate gets positive credit and one with a bad strand rate gets negative credit. If you are league average OR if you don’t get any opportunities you simply get a 0. So Kimbrel doesn’t strand runners because he doesn’t inherit them. Until he shows that he can do that, I consider him average at that part of the stat.

      I use Expected OBP (my own creation) for the not allowing runners part. This is simply how often the allow runners to reach base, with a little statistical correction that assumes some pitches get lucky on balls put in play and some get unlucky. Then I score it the same way as strand rate by comparing to league average.

      For not giving up runs, I take their average ERA and FIP (fielding independent pitching) and do the same process.

      Basically, this stat is a comparison from league average. Kimbrel does very well in the runners and runs allowed part but doesn’t get credit in the stranding runners part. He still scores well, but he doesn’t score at the very top. My goal with this stat was to create a way to measure relievers that partly measures their underlying skill and partly measures how well they pitch in certain situations. The way I would tell you to read the Kimbrel ranking is that he’s a great reliever but would be better utilized if they allowed him to come into games with men on base because he is probably going to succeed at putting out those fires.

      Don’t take it as gospel, it’s just another way of looking at relievers, who I think are pretty hard to measure a lot of the time.

      If you haven’t seen it already, we have a page explaining a lot of advanced stats. Might be helpful as you learn more about them. https://newenglishd.com/stat-primer/

3. Tigers Pull The Trigger, Grab Veras for Vasquez | New English D July 29, 2013 at 11:17 am | Reply

   […] and there aren’t any red flags involved. By New English D’s own reliever metric, SOEFA, Veras comes in at 0.24, which is an above average […]