Why Pitcher Peripherals Aren't Good Enough Anymore


Originally published at DFSEdge.

There was once a period in fantasy baseball when looking at peripheral statistics gave you a significant edge over your competition.  When most people were evaluating pitchers based on rudimentary measures like wins and ERA, being the only kid on the block looking at strikeout rate and walk rate gave you a huge edge.  It’s one of the big reasons Ron Shandler won both Tout Wars AL and NL in 1998 using his LIMA plan, which was heavily based on pitcher peripherals.

In today’s ever-evolving landscape, however, understanding peripherals merely allows you to keep up with the competition.  Looking at things like groundball rate is a necessary but not nearly sufficient condition for finding long-term success in fantasy baseball, daily or otherwise.

You’ve surely noticed in my Strong Plays articles that I make frequent mention of strikeout rate, walk rate, groundball rate, and xFIP (Expected Fielding Independent Pitching—basically combining the previous three stats into an expected ERA), but I don’t strictly adhere to what these things tell us.  For one thing, even though strikeout rate is more stable and a better predictor of future performance than ERA is, it is not perfectly stable.  We are still dealing with a finite sample size, and there is always going to be some margin of error no matter what stat we’re looking at.

And that’s really what the next step in the process is all about: deciding whether to believe the peripheral stats or not.  They’re better than wins or ERA, but they’re hardly the definitive answer when we’re trying to figure out just how good or bad a pitcher is.  Peripherals are now commonly used to tell whether surface stats are legitimate, but how do we tell whether the peripherals themselves are legitimate?

Sample Size — The first stop in the process.  How long has the player been performing at this level for?  In general terms, the longer he's been doing it, the more likely it is to be real.  In more specific terms, there are some thresholds and guidelines that I've found in previous research on when pitchers' stats stabilize that you can check out.

History — What did the pitcher do last year?  Has the pitcher done this before?  If he’s young, how did he perform at the minor league level?  These are the kinds of questions that need to be asked.  And the smaller the immediate sample size, the more important history becomes.  A 12.0 K/9 in a single game seems much less legitimate once you realize that the pitcher’s K/9 was just 5.0 in his previous 50 starts.

Pedigree — The younger the pitcher, the more important this becomes.  Was the pitcher a top draft pick or a top prospect?  The more highly regarded he was (and the more recently that was the case), the more likely his improved performance is to be legitimate.

Scouting and PITCHf/x Data — How good is the pitcher’s stuff?  Has he added a new pitch?  Added velocity?  Improved his pitch movement at all?  Is he using his pitches differently?  PITCHf/x can tell us a lot of these things now, but scouting by eye is also important, especially when it comes to things that aren’t as easily measured and analyzed, like command or pitchability.

Contextual Factors — I harp on this all the time, but context is so incredibly important.  All strikeouts are not considered equal.  Striking out Dustin Pedroia is very different than striking out Mark Reynolds.  Then you have to consider platoon factors, park factors, home field advantage, the catcher, the umpire, everything.  All of these things play a part and, in combination, have the potential to make peripherals look drastically better or worse than they ought to be.

With the season ending on Sunday, this will likely be my last daily fantasy baseball strategy article of the season, but in many ways it could have been the first.  Understanding the uses and limitations of statistics is of incredible importance for those who hope to succeed at this game.  After all, evaluating players is a huge part of the process.  If you haven’t been thinking about them in this way before, hopefully this article will start you down the path to doing so.  It’s been a lot of fun writing about daily fantasy baseball strategy at DFSEdge this year, and I hope I get to do so again next year, diving even more into the kind of topics that provide savvy daily players an edge over their competition.