On BABIP and Buying Low

As we get deeper and deeper into the season, we learn more and more about each player. This makes perfect sense, on two fronts. First, the more data we have on anyone (in any aspect of life), the better we can understand them and the more accurately we can predict behavior going forward. Secondly, we learn more and more about certain aspects (peripheral stats) of the player, which gives us an even better chance of predicting performance going forward.

For example, we know from Pizza Cutter’s work that a player’s HR/FB rate stabilizes at around 300 plate appearances. Most hitters have now hit that mark, or will very shortly. Once they hit the benchmark, we can feel much more confident when predicting home runs going forward. If you want to familiarize yourself with all of the intricacies of in-season hitting benchmarks, be sure to read Eric Seidman’s post on the subject, but below are the final results of the study:

50 PA: Swing %
100 PA: Contact Rate
150 PA: Strikeout Rate, Line Drive Rate, Pitches/PA
200 PA: Walk Rate, Groundball Rate, GB/FB
250 PA: Flyball Rate
300 PA: Home Run Rate, HR/FB
500 PA: OBP, SLG, OPS, 1B Rate, Popup Rate
550 PA: ISO

Pop quiz: What’s the biggest stat that you don’t see listed here? The answer is C) BABIP, and similarly, AVG. This means that throughout the length of one season, BABIP never really stabilizes and become reliable.

This is one of the biggest problems in fantasy leagues. When acquiring a player you believe has been unlucky thus far, you are still taking a shot in the dark. Guys have been known to have unlucky seasons, so he may not perform any better once you acquire him. Then again, he could perform much better, and you could get one hell of a steal. If I were to give advice to an owner trying to buy low on a player’s BABIP, I would always suggest doing so. If you want to increase your odds of finding a rebounding player, grab a few of them and hope that one of them works out. Of course, if you’re in a keeper league, you should buy low more often. Because, even if the player doesn’t rebound for the rest of the season, chances are he will be back to his normal self the next season, when we get to hit the “reset” button.

To put it simply, when buying low on BABIP, you just have to ask yourself one question: “Do I feel lucky?”

Well, do ya, punk?




Print This Post

Zach is the creator and co-author of RotoGraphs' Roto Riteup series, and RotoGraphs' second-longest tenured writer. You can follow him on twitter.


10 Responses to “On BABIP and Buying Low”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Matt Defalco says:

    Nice article and I really like the comparison of PA to which stats become reliable.

    Vote -1 Vote +1

  2. jimbo says:

    That’s why I’d like to see xBABIP in the stats. If I see someone with a low BABIP, but their profile is light in LD% then I’m less confident buying low. If, however, you see from the start of the season their line drives are trending up (and/or other batted balls are trending well) that’s when I think there’s a greater likelihood of a rebound.

    Earlier in the year I noticed such a trend with Beltre. Figured I had a very short window to get him and sure enough by the end of that week he was already heating up.

    BABIP can be useful IMO, but it takes digging to go from a coin flip to a well-educated guess on a player. In the end, you still have risk of no rebound though…just have to minimize it as much as possible.

    Vote -1 Vote +1

  3. Bobby says:

    I’d like to see xBABIP here also. I’ve been calculating it myself using the oft-referenced xBABIP tool, but it takes time.

    Vote -1 Vote +1

    • dexterfowlerfan666 says:

      Hear hear. It’s a tremendous pain in the Bumgarner to calculate Carty’s xBABIP, especially for people that are as terrible at Excel as I am.

      Also, is there a source that states where BABIP stabilizes over the course of a hitter’s career? 800 PA? 1,000? 1,500?

      Vote -1 Vote +1

  4. Jaybird says:

    “500 PA: OBP, SLG, OPS, 1B Rate, Popup Rate
    550 PA: ISO”

    so, if SLG is stabilized at 500PA and ISO is stabilized at 550PA, doesn’t that imply that AVG is also stabilized at 550PA since SLG – AVG = ISO?

    Vote -1 Vote +1

  5. davemcgr says:

    I’m not sure I follow this article. I understand that you cannot assume that a BABIP will necessarily stabilize throughout a single season due to a small sample size but unless plate appearances are not independent within an abnormally low/high BABIP season a player’s expected BABIP for the rest of season should be constant (and unaffected by the players current BABIP). Are you claiming that plate appearances are not independent or merely letting readers know that a rebound or regression on BABIP is far from guaranteed under a single season sample size?

    Vote -1 Vote +1

  6. Thanks for the food for thought.

    Vote -1 Vote +1

  7. Mike G. says:

    Very well and succinctly put. A player might be lucky or unlucky with his BABIP or FIP, but that doesn’t mean he’s going to necessarily reverse the trend going forward. It could just be “one of those years”.

    Vote -1 Vote +1

  8. Hello, this is my first comment. Perfect!

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *