Comments on: TOP BREWERS STORYLINES OF 2015: Brewers Bullpen and Bullpen Evaluation http://milwaukee.locals.baseballprospectus.com/2015/12/29/top-brewers-storylines-of-2015-brewers-bullpen-and-bullpen-evaluation/ Just another Baseball Prospectus Local Sites site Tue, 13 Nov 2018 16:53:42 +0000 hourly 1 http://wordpress.org/?v=4.1.1 By: Milwaukee - Baseball Prospectus http://milwaukee.locals.baseballprospectus.com/2015/12/29/top-brewers-storylines-of-2015-brewers-bullpen-and-bullpen-evaluation/#comment-5346 Tue, 05 Jan 2016 16:30:19 +0000 http://milwaukee.locals.baseballprospectus.com/?p=3048#comment-5346 […] week, BP Milwaukee looked at some of the bigger storylines from the 2015 Brewers campaign. Julien Assouline covered, in my opinion, the most notable development of all: the incredible success of the bullpen. Even […]

]]> By: Julien Assouline http://milwaukee.locals.baseballprospectus.com/2015/12/29/top-brewers-storylines-of-2015-brewers-bullpen-and-bullpen-evaluation/#comment-5192 Thu, 31 Dec 2015 13:47:06 +0000 http://milwaukee.locals.baseballprospectus.com/?p=3048#comment-5192 Thanks for the comment AD. I don’t think it’s as simple as simply inserting WPA. I contacted the stats team and PECOTA does take into account context-dependant stats (all projections systems do) for WARP. That being said, BP does not use the leverage system. One reason is, “a lot of random effects end up getting applied as something other than random by people utilizing ‘leverage'”. Some context-dependant stats have also actually been removed this year in the WARP projections.

]]> By: ADprospectus http://milwaukee.locals.baseballprospectus.com/2015/12/29/top-brewers-storylines-of-2015-brewers-bullpen-and-bullpen-evaluation/#comment-5129 Wed, 30 Dec 2015 13:49:51 +0000 http://milwaukee.locals.baseballprospectus.com/?p=3048#comment-5129 Another good article, Julien. I like the notion of teams looking for efficient ways to improve one component of their roster as an indirect means of compensating for a weakness elsewhere. I’ve been looking at that in the context of the Tigers’ bullpen for a couple seasons now (e.g., https://aldland.wordpress.com/2015/07/10/window-shopping-pigs-in-the-pen/), and I think it’s a simple but very important idea.

On the reliever WPA/projection discussion, I wonder if it’s as simple as the projections simply failing to consider reliever WPA– a non sequitur of sorts– rather than something to do with a correlation connection. In other words, if PECOTA projections included reliever WPA (I’m assuming they don’t), the positive relationship you find between teams beating their PECOTA projections and WPA might not exist. (Perhaps the fact that that relationship does exist, together with the more substantive arguments Carleton and Sullivan highlight, suggests that PECOTA and other projections ought to be using WPA to judge relievers.)

]]>