Before we get started, we want to make sure everyone knows our publisher is making the Philadelphia Phillies chapter of Baseball Prospectus 2016 freely available as a PDF! Go get your copy now by clicking here.
We posted last week about some of the improvements we've made to Baseball Prospectus 2016. We want to keep these posts manageable in length, and there were a lot of changes, so we'd like tell you about some of the improvements and additions to the Baseball Prospectus 2016 stats and projections packages.
J.P. Crawford is #4 on Baseball Prospectus' Top 101 Prospects of 2016.
Pitcher, hitter, and catcher PECOTAs include two years of projection for every player.
The tendencies of baseball fans are heading to extremes. On one hand, daily fantasy leagues are extremely popular and heavily-brovertised. On the other, fantasy owners and fans of specific organizations alike want to track the stars of tomorrow earlier and earlier, over longer terms. Some of the extra heft in Baseball Prospectus 2016 supplies data which caters to the latter group: we've added a second season of projections to the player boxes in the book.
As a disclaimer, of course we don't really know what players will do in 2017 with certainty. As Nate Silver wrote back in 2003:
As far as I am aware, no other forecasting system has published a complete set of projections that run more than one year into the future. Apart from the mathematical limitations of competing models, there isn't much good reason for this: There are many situations in which it's markedly more helpful to look several years into the future than to take a piecemeal approach, whether it's determining which player to lock up for his arbitration years, or deciding who to pick up in a keeper league.
Certain problems tend to compound over time. Nagging injuries become chronic, and then debilitating; eroding foot speed produces ever-lower batting averages; the stress on a pitcher's arm accumulates without warning, until, like an earthquake fault, it snaps. The further one projects into the future, the more uncertainty there is.
Nate Silver, PECOTA Five-Year Forecasts
What we do know are statistical likelihoods, and the PECOTA model offers as much certainty as is available. As Nate pointed out later in that same article, baseball is a game of attrition. Young players don't always grow. Nascent superstar Carlos Correa is projected to show no improvement over his Rookie of the Year season through 2017. And while there isn’t much about the Cubs that’s a cause for worry, PECOTA shows Javier Baez as likely to perform at a pedestrian league-average level for the next two seasons. Meanwhile, Ruddy Giron of the San Diego Padres is projected to be only a .228 TAv hitter in 2016 but grow into an near-league-average hitter with a .256 TAv the following season. As it has always been, PECOTA is influenced by the career trajectories of comparable players, with the "shape" of a player's statistics being more important in determining similarity than the "size" (so players are often compared to much better or worse players–overall–simply because their strengths and weaknesses align). Giron's top three comparables are the impressive trio of Addison Russell, Carlos Correa, and Manny Machado. We hope you enjoy the extra year as much as PECOTA predicts Giron will.
New Catcher Statistics
The past eight years has seen enormous progress in the type of data available, from detailed pitch-by-pitch results back to 1988 and full pitch flight tracking via PITCHf/x and Trackman since 2008. New techniques for analyzing data have also been applied to questions new and old. We feel the updates in Baseball Prospectus 2016 are among the most exciting–and expansive–we’ve ever released.
In Baseball Prospectus 2016 catchers have their own stat block for our new catcher-specific defensive metrics. You'll see the number of pitches, framing, blocking, throwing, and total runs above average split out across all catching lines.
J.P. Arencibia has had many addresses over the last three years.
Baseball Prospectus has developed an array of new metrics that are used not just on their own, but are included in FRAA as and, by extension, WARP. Catcher defense now includes pitch framing (Called Strikes Above Average) and pitch blocking (Errant Pitches Above Average). Our newly released throwing metrics covering both the prevention of attempts (Thefts Reduced Above Average) and the ability to throw out runners once they have gone (Swipe Rate Above Average) have also been moved into FRAA, along with an improved accounting of outfield assists.
The improved throwing data–both catching and outfielding–and new blocking model goes back to the 1950s for Major League Baseball. We have Major League framing as far back as 1988, and our more precise model based on pitch-tracking data is applied from 2008 onward.
We've also added Minor League coverage for our catching statistics. The more traditional metrics around throwing and blocking go back as far as 2006, and we also provide framing numbers for Triple-A (starting in 2006), Double-A (Texas League in 2008, full coverage by 2012) and the 2015 New York-Penn League.
Finally, we added the new catching metrics to PECOTA and have projected them for 2016 and 2017. You can find a sample of the projections at the bottom of the J.P. Arencibia catcher stats above, and they are projected for every catcher we have a PECOTA projection for in Baseball Prospectus 2016.
Pitching Stats Changes
Pitcher statistics include DRA, CFIP, a reformulated DRA-based pitcher WARP, and 95th percentile pitch MPH.
We've switched over to a new pitching metric, Deserved Run Average, for Major Leaguers (MiLB DRA coming soon, the book uses RA/9). We've written a lot about DRA over the last year. DRA is included in WARP but we are not including it in PECOTA. Why? DRA is designed to describe what happened, not project what will happen. For PECOTA we are using another new stat, cFIP, or "contextual FIP". cFIP is shown for Major and Minor league pitchers, for past seasons and the two seasons of PECOTA included.
We've added pitcher's "Peak Velocity" for all Major Leaguers. This is based on the 95th percentile of pitch speed for the entire season's collections of pitches thrown by each pitcher. We hope this provides some insight into what kind of reliable upper-end speed a pitcher has reached.
We've updated the errata list for the book; head on over there if you see any issues and leave us a comment and we'll add it to the list. Thanks!