This is our old blog. It hasn't been active since 2011. Please see the link above for our current blog or click the logo above to see all of the great data and content on this site.

Bloops: Beware the ‘Bases Fallacy’

Posted by Neil Paine on March 22, 2011

Over at The Book Blog, Tangotiger offers another friendly reminder about the required reading out there regarding the dreaded Bases Fallacy. Although it can happen to the best of us (one of my most embarrassing moments to be sure), once you know about the bases fallacy, you must do everything in your power to prevent the bases fallacy from happening again.

17 Responses to “Bloops: Beware the ‘Bases Fallacy’”

  1. Dr. Doom Says:

    Quality, Neil. Thanks. And don't worry about committing the Bases Fallacy. We've all been there.

  2. Chuck Says:

    I don't see any reference to you on the Baseball Fever blog..

  3. John DiFool Says:

    What would be interesting is a stat which tracks how many bases an offensive event generates-for example, a home run can be worth from 4-10 bases (a grand slam = the latter), while a stolen base is of course just worth one. No it still wouldn't be perfect, but it would be informative.

  4. Devon & His 1982 Topps blog Says:

    @ #3 ...if I understand you correctly, you're talking about the same thing linear weights is.

  5. Neil Paine Says:

    #2 - In case it wasn't clear, I was the sap who started that thread.

  6. Chuck Says:

    You're davis?!

    Oh, snap, the ghost of ArmchairGM lives on.....

  7. Neil Paine Says:

    Kelsdad, I presume... 🙂

  8. Chuck Says:

    You presume correctly.

    How's it going, buddy..good to see (read?) you again..

    I knew you were here, but I didn't know you were a big Kahuna.


  9. Jared Says:

    Not to be the contrarian in the group, but sometimes a fallacy isn't really a fallacy. Now, not to cause confusion, I'm not arguing that a basic measure of accumulated bases is the best way of demonstrating someone's offensive value or that all accumulated bases are created equally. The simplest way to define fallacy is "false belief". I would agree that if someone believes that it is the best offensive measure or that all accumulated bases are equal, then it would be a fallacy. However, short of that, I think we run the risk of overdiagnosing the fallacy. I think most would agree that not all hits are created equal. A single with a runner on third is more valuable than a single with a runner on first, etc. However, we use batting average as a good index of the success of a hitter, and batting average treats all hits like they're equally valued. So, I could have a hitter whose total BA is .279, but whose BA with runners in scoring position is .321. This hitter would be more valuable to his team than a hitter whose BA is .321 but who hits .279 with runners in scoring position. However, no one talks about the batting average fallacy. Batting average has hung around because it is a good general index of the success of a hitter - maybe not the best index - but a good one nonetheless. I think the same is true for accumuluated bases. If you look at the formula, in general players who have more total bases, walk more, steal more bases, get caught stealing less and do it in fewer plate appearances than their peers are going to be more productive players. So, to make a short story long. Don't throw the baby out with the bathwater. Sometimes the use of this or similar statistics can be used without be fallacious.

  10. Noodle Says:

    But then aren't you saying the Slugging pct is flawed too? Treating a HR as 4* a single when it can be worth much more? And if so, is OPS flawed sicne Slugging in one of its two metrics?

  11. Neil Paine Says:

    #10 - OPS is a flawed metric -- everyone just uses it as shorthand out of convenience, because it's easy to calculate and correlates well with run-scoring. It's the most meaningful of the simple stats... or the simplest of the meaningful stats.

    But if you have a chance to calculate, say, wOBA, you do that instead of using OPS.

    And that's the point: people aren't re-inventing this bases per out wheel because they think it's the next shorthand stat (it's not nearly simple enough, anyway). They do it because they think it's the next "holy grail" stat, without realizing that linear weights A) is no more difficult to calculate, and B) values offensive events much more accurately.

    #8 - Great to see you around these parts, KD. (For those wondering, Chuck & I used to write for a site called ArmchairGM a long time ago.)

  12. Chuck Says:

    Ah, the good ol' days.

    I've been around, writing primarily at, Mike Pagliarulo's old site.

    More baseball, less stats, and you know that's always been my preference.

    Still working with Dave Smith at Retrosheet, so the connection to here has always been unbroken.

    Still think your "100 Greatest QB's" article back in the day is still one of the best pieces of work I've ever seen....anywhere.

    I hope your original opinion of Tom Brady remains the same...:)

  13. Neil Paine Says:

    Eh, sort of... he did go out and have two of the best QB years of all time in the four seasons since that post. 🙂 Eventually all of us stat-heads had to eat crow on Brady to some degree.

  14. Neil Paine Says:

    I say this as a Patriots fan, btw. It's never easy when the stats say one of your heroes is overrated -- a feeling I would imagine every saber-minded Derek Jeter fan knows intimately. Or saber-minded Kobe Bryant fans. Or saber-minded Derrick Rose fans. Or saber-minded Carmelo Anthony fans. Or-- OK, I'll stop now.

  15. Chuck Says:

    Having to eat crow isn't limited to saber-heads.

    I've eaten more than my share, and for sure I'm not done yet.

  16. Khalilah Rohleder Says:

    I was more than happy to find this web-site.I wished to thanks for your time for this wonderful learn!! I definitely enjoying each little little bit of it and I've you bookmarked to take a look at new stuff you weblog post.

  17. John Autin Says:

    Neil, thanks for the required reading pointers. I read the one from Hardball Times titled "The great run estimator shootout (part 1)". (Dated 2009, but ... still news to me!) One thing I learned there: "The relationship between runs scored and OPS relative to average is about 2:1, and this is why a lot of run estimator studies underrate OPS; they don't take that into account when translating OPS into runs." I hate to say it, but I hadn't directly pondered the relationship between runs and OPS.

    But a caveat to anyone who follows that link: There's no link to part 2. (I just now emailed the author, Colin Wyers, to ask what's up.)