Baseball Prospectus is looking for a Public Data Services Director. Read the description here.

I don’t take advantage of this column very often for mere venting.
(Some would correctly argue I could have left of “for mere venting”
in the preceding sentence and still be accurate). But sometimes you
just want to share your frustrations. I don’t know how illuminating
this will be, but it’s good for me to get it out there anyways.

I’ve been working behind the scenes to make our
stat reports
reconcile better with sources such as ESPN, particularly in
less official categories such as holds and blown saves.

Accuracy is relative though–should I try to be correct according to
the letter of the rule? The spirit of the rule? Or the way it’s done
in practice? None of those are exactly the same, as you’ll see below, as I
list Five More Reasons to Hate the Hold.

  1. Two identical performances in virtually identical game situations can result in one
    getting credit for a hold, and the other not.

    Consider the following games:

    Game A: Ricardo Rincon comes in on August 1st, 2004, with a 3 run lead,
    one out, bases empty, records one out before being removed, allowing no runs to
    score, and bequeathing no runners.

    Game B: Willis Roberts comes in on August 31st, 2001 with a 3 run lead,
    one out, bases empty, records one out before being removed, allowing no runs to
    score, and bequeathing no runners.

    Rincon gets credit for a hold, but Roberts doesn’t. Why?

    Answer: Because in game A, Rincon came in in the 8th inning, and thus had the
    opportunity to pitch more than one inning with a 3 run lead, creating a
    save situation. In game B, Roberts came in with 1 out in the 9th. Since Roberts
    wouldn’t have the opportunity to pitch 1 full inning, and because the
    tying run was not on base, on deck, or at bat, it’s not a save situation.

    It’s not that he actually pitched 1 inning in either game, but that he
    could have been left in long enough to record a save. This goofiness is
    inherited from the definition of a save–3 run lead and 1+ IP, or any
    lead with tying run on deck or better. Thus with a 3 run lead and no one
    on base, you can’t get a save (or a hold) with 1 out in the 9th, but you
    can with 1 out in the 8th, while pitching exactly the same way in the
    same base-out and score differential situations.

  2. Holds are sometimes awarded in direct contradiction to the rules spelled out for them.

    According to ESPN’s glossary,
    a pitcher has to come into a save situation in order to get a hold. Part
    of the definition of a save situation is that the pitcher is not the winning pitcher of record.
    This should make it impossible for a pitcher to earn a hold and a win in the same game.

    Yet Guillermo Mota got credit at ESPN for both a win and a hold on
    July 3rd, 2004.

  3. Pitchers can get more credit for not finishing a game than for finishing it.

    A pitcher awarded a hold can’t be the pitcher who finishes the game, even
    if he can’t be awarded a save.

    Using the 2004-07-03 game mentioned above, Mota pitched the 7th and
    8th, and gave way to Eric Gagne in the 9th. Mota gets the win (he was the most
    effective pitcher in the scorer’s discretion) and the hold (whether he
    should have or not, see #2 above), Gagne gets the save.

    Hypothetically, if Mota had pitched the ninth, then neither a hold nor
    a save would have been awarded. According to the hold rule, he would still
    have gotten the win, but not a hold, because he finished the game. And since
    the winning pitcher can’t also get a save, no save would have been awarded.

    Net effect:

    Pitch 2 IP — get a win and a hold, and the next guy goes 1 IP and gets a save
    Pitch 3 IP — get a win, but no hold and no save (for anybody).

    Mota got more “credit” for not finishing the game (a win and a hold)
    than he would have if he had finished the game (just a win).

  4. Relievers apparently can’t get holds or blown saves if they come in
    before the 6th inning.

    Example, on June 14th, 2002,
    Ron Villone came in to relieve Kris Benson
    with no one out in the 4th inning and a two run lead. He pitched three
    innings, allowing two baserunners and no runs. He qualified for a
    save in all three ways listed in the save definition (came in with
    the tying run on deck, came in with a 3-run lead or less and
    pitched at least one inning protecting the lead, and pitched
    three innings effectively while preserving the lead), but
    did not get credit for a hold–or the win, as the Pirates lost
    in 11 innings.

    Ironically, the three pitchers who came after him combined to
    give up four baserunners and two runs in two innings of work,
    “earning” three holds among them.

    PITTSBURGH                   ip       h   r  er  bb  so  hr    era
    K Benson                      3       1   0   0   1   1   0   7.42
    R Villone                     3       1   0   0   1   5   0   5.49
    M Lincoln (H, 4)                1/3   2   1   1   0   1   1   0.81
    S Sauerbeck (H, 15)             2/3   0   0   0   0   2   0   2.51
    M Fetters (H, 10)             1       1   1   1   1   1   1   2.13
    M Williams (L, 1-1; B, 2)     2       4   2   2   1   2   2   2.13

    What happened here?

    If the starter goes 5 IP, and a reliever pitches the last 4 IP holding the lead,
    the reliever get a save (pretty much regardless of the lead, thanks to the 3-inning
    of effective pitching save rule). That’s entering to start the 6th, but not coming in
    before the 6th. If the starter doesn’t go 5 IP, then the starter can’t get the win.
    If the reliever then finishes the game, and the reliever has to be credited with a
    win, which makes him ineligible for a save. Hence coming in before 5 IP are
    finished make it not a “save situation”. Which means it’s not a “hold situation” either.

    So if a pitcher comes in with 2 out in the fifth and the tying run on base,
    and retires the batter, there’s no hold. But if he does the same thing
    with 2 out in the seventh and the tying run on base, he gets a hold.

    I don’t know what happens if the starter pitches to one batter, gets hurt,
    and the next pitcher pitches 4.7 innings, and the reliever pitches the last
    4.3. The 1st reliever can get credit for the win, since he’s not the
    starter. So I guess the 2nd reliever could get a save. And in that
    situation, the 2nd reliever would be entering the game in the 5th, but
    probably not get the win. Whether he could get a hold or a save isn’t
    clear to me.

  5. This is more related to the hold’s sister stat, the blown save, but we’ll
    include it in my rant just the same.

    The date is June 12th, 2004, and
    Pittsburgh has a four run lead heading into the 8th inning over Oakland
    . Jose Mesa
    comes in to start the inning in relief of the ejected Salomon Torres. He strikes out
    the first hitter, Damian Miller. Then the trouble starts. Home run, single,
    single, single, walk, sac fly, single, double, and Mesa is pulled. He
    gives up six hits and a walk in two-thirds of an inning. Though the loss is
    later pinned on Mark Corey, it was Mesa’s pitching that did the most damage.
    Mesa is not charged with a blown save, because the situation he came into was
    not one where he could have earned a save or a hold.

    It also seems to me that if a reliever comes in with a four run lead, and
    blows it by giving up 5 runs (and all cleanly in Mesa’s case–no inherited
    or bequeathed runners were involved in the scoring), he should get charged
    with something. But he doesn’t get the loss, because his team rallied to
    tie the game in the 9th before losing in 11 innings. And he can’t get a blown
    save, because it wasn’t a “save situation” when he entered the game (his team was
    ahead by too many runs). Not a blown save, but a blown lead nonetheless.
    How silly is it that there’s a name for surrendering a small lead, but not a large one,
    which is presumably more damaging and reflects even worse performance than a blown save?

    Anyone want to come up with a name for a stat reflecting “big leads squandered”?

Thanks for letting me vent. I feel much better now.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe