Me vs. reality: 2012 edition

This year, as is my annual tradition, I predicted how the BBWAA Hall of Fame vote will go. I don’t just say in/out, but give actual predicted percentages. Well, the vote is out, so how did I do? Let’s see.

Below is a chart comparing the actual BBWAA vote totals versus my predictions for the 2012 elections – plus the difference between the two (saying how high/low my predictions were compared to reality).

Name	     BBWAA	Me	Diff.
Barry Larkin	86	82	  -4
Jack Morris	68	65	  -3
Jeff Bagwell	56	54	  -2
Lee Smith	51	52	   1
Tim Raines	49	52	   3
Edgar Martinez	37	39	   2
Alan Trammell	37	32	  -5
Larry Walker	23	27	   4
Mark McGwire	20	24	   4
Fred McGriff	24	24	   0
Dale Murphy	15	19	   4
Don Mattingly	18	18	   0
Rafael Palmeiro	13	15	   2
Bernie Williams	10	12	   2

14 guys predicted, and I was within five percentage points with all of them. I was off by an average of 2.6 percentage points. Yeah, I like that.

In all:

Guessed perfectly twice (Don Mattingly, Fred McGriff).

Off by one percentage point once (Lee Smith).

Off by two percentage points four times (Jeff Bagwell, Edgar Martinez, Rafael Palmeiro, Bernie Williams)

Off by three percentage points twice (Jack Morris, Tim Raines)

Off by four percentage points four times (Barry Larkin, Larry Walker, Mark McGwire, Dale Murphy)

Off by five percentage points once (Alan Trammell).

That’s another year I can claim to know what I’m talking about.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Barry Larkin career highlights
Next: A baseball card mystery: Alan Trammell and George Vukovich »


  1. Chris Jaffe said...

    Via VC – sure.

    Via BBWAA – well, he has a shot, but I don’t think they’ll vote him in.  Too many guys entering the ballot in the next year or two.

  2. Tangotiger said...

    Chris: the fair thing to do is to REMOVE all the posted ballots when comparing the final numbers to your forecast.

    So, if you are aware that 70 of 120 posted ballots had Raines, then your forecast should remove the 70 and the final number should remove the 70, and then you compare.

  3. Chris Jaffe said...

    Tango – first a nitpick.  When I wrote the column only 70 or so ballots were known to repoz, not 120. 

    Second, I really don’t use the tally for Raines or other backloggers.  I used to use it more for them, but I’ve found that the other guidelines I use (strength of ballot, over the top surge, comparable candidats, etc) do a better job predicting things than the tally.

    I use the tally pretty much just for newbies.  This year that was just Williams – and even there I took some liberties.  I knew the tally was always well off on Mattingly, so I assumed it would be on Williams as well.  So I guessed well north of reality for him.

    At any rate – and most importantly –  I don’t see why I would remove it.  The tally is one tool I use to get the results.  I’ve never claimed to be independent of it.  In my annual prediction column I highlight it as one of my ten guidelines. 

    I’m trying to predict how the total vote would go, not the missing 75% repoz doesn’t predict.  If you want you can always work backwards from my predictions and repoz’s tally and figure how I did predicting the others, but I’m not sure what purpose that would serve.

    Like I said, though, at this point I just look at the tally to figure out newbies.  Backloggers I handle on my own.  Maybe that’ll change next year with the utter nightmare coming in, but I’ll worry about that next year.

  4. David said...


    I totaled the number of votes, and got 2921, on what I believe were 573 ballots.  That works out to 5.09 names/ballot, which would be, according to your last article, the lowest all-time, and by nearly a quarter of a name.  Is that correct, and, if so, why do you think that is?

  5. Chris Jaffe said...

    Tango – I think I see what you mean by baseline.  Going by repoz, here’s the gap between his tally and results for the same 14 guys:

    Perfect: McGriff, Trammell

    Off by 1 percentage point: Bagwell, Palmeiro

    Off by 2 percentage points: McGwire

    Off by 3: Murphy, Larkin, Raines

    Off by 5: L.Walker, E.Martinez

    Off by 6: L.Smith

    Off by 7: B. Williams.

    Off by 8: Mattingly

    Off by 9: Morris

    Average margin of percentage points off by: 3.8.

  6. Tangotiger said...

    Excellent!  So your expertise is able to spot the bias to reduce the average error by 1.2 percentage points, compare to a non-random sample of 26% submitted ballots.

  7. Tangotiger said...

    By the way, it looks like the people who voted for Morris didn’t declare that they voted for Morris.

    87 of 148 publicly declared their Morris intention (59%).

    That means that 295 of the 425 non-public ballots voted for Morris (69%).

    So, they are either scared, or part of the old group that has no desire to state their choices.

  8. Chris J. said...

    David – I don’t have the exact numbers in front of me but I think it rounds up to 5.10 names/ballot, and yep – it’s the all-time low by a full quarter of a name per ballot.

    Why?  There’s a weak rookie crop and the BBWAA wasn’t very thrilled with the backlog.

  9. Tangotiger said...


    “14 guys predicted, and I was within five percentage points with all of them. I was off by an average of 2.6 percentage points. Yeah, I like that. “

    So, you have to give that context.  If you went strictly with the Repoz or that other person I found, that is your baseline.

    I don’t know what off by 2.6 points means, if we have say a sample of 20% of the ballots to compare against.  How much were THOSE off by?

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>