Aim For The Head: Mailbag: Outcomes and Outrages

January’s recap of the 2005 Three True Outcome leaders generated several reader questions, though only one came up repeatedly:

Why not include HBP as a true outcome? If you do, Dunn beats out Bellhorn. I know FTO doesn’t look as good as TTO, but I want to see Dunn win. Actually, Jason Giambi also sneaks in ahead of Bellhorn.

–J.S.

Isn’t there a fourth True Outcome – HBP? Much like BB it’s a batter’s skill as well as a pitcher’s deficiency as well as being defense independent. Because HBP is relatively uncommon, it might make sense to change BB to BB or HBP as a single Outcome.

–M.S.

Why aren’t HBPs considered part of the equation? After all, when someone like a Craig Biggio gets plunked 20, 25, 30 times a year or so, doesn’t that have a substantial effect on his TTO%?

–C.H.

How come hit by pitches are not considered True Outcomes? They certainly keep the ball away from the defense, after all. Would any of the TTO standings change if HBPs were counted as BBs? Or am I a heretic for even asking such a thing?

–C.K.

Shouldn’t HBP’s count as a 4th true outcome? Just wondering if Giambi (19) or A-Rod (16) might have been higher in the rankings if these were counted. Or are they already being counted?

–J.P.

This is actually a question that gets asked every year, when we discuss Three True Outcomes. Yes, in principle an HBP could be considered a True Outcome. It’s tradition as much as anything else that makes the formulation the major events (HR, BB, SO) rather than minor ones like HBP. The standard definition of the Three True Outcomes date back to at least the mid-1990’s on the Usenet newsgroup rec.sport.baseball when the “Rob Deer Fan Club” first arose. Baseball Prospectus adopted the tongue-in-cheek statistic, and we’ve chosen to leave the classic definition intact.

What is the practical application of TTO in player analysis, development and valuation? How would an MLB general manager, for instance, use this information in decision-making regarding acquisition and retention of roster talent?

–C.H.

There really isn’t any practical application, at least for batters (though for pitchers, the entire defense independent pitching edifice is built upon it). For batters, I’ve presented research that’s shown that batters do have the ability to affect the results of balls in play, so focusing on just the TTO would ignore a significant aspect of their value. There may be some predictive power in TTO for minor leaguers, but I’ll leave that for Nate Silver to incorporate into PECOTA.

Why rank according to NLST? Seems like that is the only reason Sexson is close to Adam Dunn. Sure he struck out more in 15 fewer plate appearances, but Dunn had 30 more walks than Sexson! What’s wrong with taking the total of the 3 normalized rankings, or the average? You’re ranking according to the players’ worst of the TTOs.

–A.F.

I explained the rationale for this a little more throughly in a previous year‘s TTO column. The basic idea is that in crowning a TTO champion, we want to reward someone who excels at each of the individual Outcomes. Otherwise, you end up with a player like Barry Bonds who excels in hitting homers and walking, but rarely strikes out. His strength in the two former categories is so dominant that averaging the three scores would put him ahead of players who actually strike out, walk, and hit home runs at prodigious rates. We want the player with the strongest “weakest link.”

I enjoyed your 1/24 article on the 2005 TTO batting champ. What I’d really like to see, though, is an article on the TTO pitching champ.

–Dave Feinstein

A splendid idea. Let’s look at the pitching leaderboard:

First, we’ll list the top 20 pitchers (min 300 batters faced) in raw TTO%

NAME              PA  HR   BB   SO  TTO TTOR    RK   NHR   NBB   NSO  NAVG  NLST   RK
Ugueth Urbina    330  12   39   97  148 44.8%    1  1.35  1.45  1.79  1.53  1.35    1
Michael Wuertz   319   6   40   89  135 42.3%    2   .70  1.54  1.70  1.31   .70  126
Brian Fuentes    321   6   34   91  131 40.8%    3   .69  1.30  1.72  1.24   .69  131
Oliver Perez     471  23   70   97  190 40.3%    4  1.81  1.82  1.25  1.63  1.25    2
Mark Prior       701  25   59  188  272 38.8%    5  1.32  1.03  1.63  1.33  1.03    7
Fra. Cordero     302   5   30   79  114 37.7%    6   .61  1.22  1.59  1.14   .61  160
Andrew Sisco     329   6   42   76  124 37.7%    7   .68  1.56  1.40  1.22   .68  142
Scot Shields     375   5   37   98  140 37.3%    8   .50  1.21  1.59  1.10   .50  192
Juan Rincon      319   2   30   84  116 36.4%    9   .23  1.15  1.60  1.00   .23  216
Daniel Cabrera   716  14   87  157  258 36%     10   .73  1.49  1.33  1.18   .73  112
Ryan Dempster    401   4   49   89  142 35.4%   11   .37  1.50  1.35  1.07   .37  207
Jason Schmidt    757  16   85  165  266 35.1%   12   .78  1.38  1.33  1.16   .78   79
Jake Peavy       812  18   50  216  284 35%     13   .82   .75  1.62  1.06   .75   99
Scott Kazmir     818  12  100  174  286 35%     14   .54  1.50  1.29  1.11   .54  178
Seth McClung     501  20   63   92  175 34.9%   15  1.48  1.54  1.12  1.38  1.12    5
Doug Davis       946  26   93  208  327 34.6%   16  1.02  1.20  1.34  1.19  1.02    9
Rob Tejeda       371   5   51   72  128 34.5%   17   .50  1.68  1.18  1.12   .50  190
Carlos Zambrano  909  21   86  202  309 34%     18   .86  1.16  1.35  1.12   .86   43
Aaron Heilman    439   6   37  106  149 33.9%   19   .51  1.03  1.47  1.00   .51  186
Brett Myers      905  31   68  208  307 33.9%   20  1.27   .92  1.40  1.20   .92   28

Next, the 13 pitchers who were above average in surrendering HR, BB, and SO.

NAME              PA  HR   BB   SO  TTO TTOR    RK   NHR   NBB   NSO  NAVG  NLST   RK
Ugueth Urbina    330  12   39   97  148 44.8%    1  1.35  1.45  1.79  1.53  1.35    1
Oliver Perez     471  23   70   97  190 40.3%    4  1.81  1.82  1.25  1.63  1.25    2
Yhency Brazoban  317  11   32   61  104 32.8%   29  1.29  1.24  1.17  1.23  1.17    3
Chris Capuano    949  31   91  176  298 31.4%   43  1.21  1.17  1.13  1.17  1.13    4
Seth McClung     501  20   63   92  175 34.9%   15  1.48  1.54  1.12  1.38  1.12    5
Ted Lilly        566  23   58   96  177 31.3%   44  1.51  1.26  1.03  1.27  1.03    6
Mark Prior       701  25   59  188  272 38.8%    5  1.32  1.03  1.63  1.33  1.03    7
Scott Downs      407  12   34   75  121 29.7%   58  1.09  1.02  1.12  1.08  1.02    8
Doug Davis       946  26   93  208  327 34.6%   16  1.02  1.20  1.34  1.19  1.02    9
Casey Fossum     725  21   60  128  209 28.8%   69  1.08  1.01  1.07  1.05  1.01   10
Barry Zito       953  26   89  171  286 30%     57  1.01  1.14  1.09  1.08  1.01   11
J.P. Howell      328   9   39   54  102 31.1%   46  1.02  1.46  1.00  1.16  1.00   12
Jose Contreras   857  23   75  154  252 29.4%   61  1.00  1.07  1.09  1.05  1.00   13

Should we have separate categories for TTO champions in relief and starting pitching? Sure, why not? As they are the only two pitchers with TTO% above 40% and are above average in each component Outcome, your 2005 TTO pitching champions are:

Starter: Oliver Perez
Reliever: Ugueth Urbina

Another recent Aim For the Head, Five More Reasons to Hate the Hold, generated its own share of reader feedback, particularly in response to the question raised in the article: “Anyone want to come up with a name for a stat reflecting ‘big leads squandered’?”

how about calling the act of ‘big leads squandered’ the Mesa.. he seems to do it every-other appearance anyways so he’s bound to own the trademark.

–D.B.

I think you answered your own question — it should be called a Mesa.

–Steve Matuszek

Call it a Mesa!

–B.C.

Great article. You missed the most obvious name for the stat for a big blown lead – the ‘Mesa’.

–J.P.

Perhaps that stat for ‘big leads squandered’ could be called the Mesa, or the Lima.

–G.H.

Quite frankly, I think you already have your name for blowing a big lead: a mesa. Ok, maybe it should more appropriately be called a Kolb, but in honor of the way in which this issue was called to our attention I think we should stick with ‘mesa’. And, I’d be interested to know who led the league in mesas last year. Might it be The Man Himself?

–G.F.

I had a good laugh watching the responses roll into my email inbox, as over half the suggestions were to call a big blown lead a “Mesa.” However, I’m not a big fan of naming a stat after a person, particularly when a more descriptive name can be found. And there were some other, non-Mesa suggestions:

I think the ‘big league squandered’ stat should be called the ‘choke’. It fits in well with ‘hold’, as in ‘choke hold’.

–Anonymous

BLOW – Big Lead Ostentatiously Wasted

–S.D.

Instead of ‘big lead squandered,’ why not just ‘blown win’ for anything that is not a ‘blown save’ situation (i.e., a lead blown where there is not a save situation, regardless of inning and regardless of whether or not the pitcher’s team wins or loses the game). A ‘BW’ would appear in the box score.

–E.D.

It’s already been created: DFA. Designated for assignment.

–Ryan

Toledo – as in, you’re going to find yourself in Toledo tomorrow.

–S.W.

DFA and Toledo definitely have humor value, but aren’t descriptive enough to make the cut. “Blown win” has possibilities, but distinguishing “blown saves” from “blown wins” doesn’t seem quite right. A blown save is a kind of blown win. Big Lead Ostentatiously Wasted (BLOW) captures the idea, although the acronym would create confusion with the existing “blown save” nomenclature, not to mention its more risque interpretations. Of these, I think I prefer calling the big-lead-squandered a “choke.” This also has another benefit. Since the concept of a hold and blown save are so screwy, we could use this as a jumping off point to define a new system of names to characterize relief appearances, with some better thinking behind it. For example, rather than a “blown save,” there’s a natural counterpart to the “choke” for a blown-save-like event called a “cough,” as in “coughing up the lead,” yet “coughing” is not as severe as “choking.” I may develop this to a probably-absurd level in another column.

Thanks for illustrating all the deficiencies with the hold rule. Your article reminded me of the July 18-19, 2001 game where the Diamondbacks-Padres game was stopped after 2 innings due to a power failure, with Schilling leaving with a 1-0 lead. Randy Johnson came in to pitch and threw 7 shutout innings and struck out 16 (a record for a reliever). He got credit for the win, but it’s amusing to think throwing 7 innings of shut out ball as a reliever wouldn’t qualify as a hold.

–Matt Knight

Believe it or not, I think it’s actually impossible for a a reliever to pitch 7 shutout innings, and get a hold. If the game is a typical 9 inning affair, then a pitcher like Randy Johnson in the game above would have had to come in with no one out in the 3rd inning which, as we’ve seen, makes him ineligible for a hold since the starter could not record the win. If Schilling had been pulled in the first inning, and Johnson actually replaced a reliever in the 3rd, in theory the first reliever could have gotten the win, though the official scorer is most likely to award it to the 7-inning 0-run performance.

If a game goes into extra innings, and a pitcher throws 7 shutout innings extending beyond the ninth, he can certainly get the win. But he could not have come in with the lead, because the game would have to be tied at the end of 9 innings, meaning that the pitcher surrendered the lead during his appearance. He could have allowed an inherited runner to score, and have no runs actually charged to him, but that inherited runner scoring blows the lead, which would mean no hold in any event.