Aim For The Head: Three True Outcomes, 2003

Periodically, Baseball Prospectus pays homage to the “Three True Outcomes” and those players who excel at creating them. A long-time inside joke at rec.sport.baseball, discussion of the Three True Outcomes (or TTO) has appeared on the pages of BP for years.

In short, the Three True Outcomes are plate appearances that end with events that do not involve the fielders: the home run, the walk, and the strikeout. Somewhat ironically, the TTO have gained prominence in recent years with Voros McCracken’s controversial (and oft-misstated) theory that pitchers do not differ significantly from each other on their ability to prevent hits on balls in play; thus making their primary differentiators of value the rates of strikeouts, walks, and home runs they allow.

But the Three True Outcomes are, at their core, a celebration of hitters, epitomized by the patron saint of the TTO, and the prototype for early BP book covers, Rob Deer. With that in mind, we start with a list of the top hitters for 2003, according to the percentage of their plate appearances that ended with a True Outcome (with a minimum of 300 PA).

 
NAME            PA      HR      BB      SO      TTO%    RANK
Thome_Jim       698     47      111     182     48.7%   1
Dunn_Adam       469     27      74      126     48.4%   2
Edmonds_Jim     530     39      77      127     45.9%   3
Bonds_Barry     550     45      148     58      45.6%   4
Giambi_Jason    687     41      129     139     45.0%   5
VanderWal_John  374     14      46      104     43.9%   6
Wilkerson_Brad  600     19      89      155     43.8%   7
Bellhorn_Mark   306     2       50      78      42.5%   8
Sosa_Sammy      589     40      62      143     41.6%   9
Hernandez_Jose  571     13      46      177     41.3%   10

Atop the list is a classic TTO player, Jim Thome. Thome is well known for his ability to keep the ball away from the fielders. He has tremendous power, walks frequently, and strikes out prodigiously.

However, not all players on the list fit this profile. Look at #8, Mark Bellhorn, who hit a scant two homers. He made the list thanks to his high walk and strikeout rates, but he’s not a TTO hitter in the mold of Rob Deer.

Part of the problem is that strikeouts (and walks) are far more common than home runs, and being exceptionally high here can offset a low homer rate. Since each strikeout counts as much as a home run, a player with no homers and 160 strikeouts looks the same as a 50 HR/110 SO player, even though the latter conforms much more closely to the ideal of a TTO hitter.

One way to correct for the magnitude of differences in raw totals is to normalize them to the same scale. We can divide the rate of each event by the major league-average rate, and express each component of TTO by the percentage by which the player exceeds the MLB average. This is similar in concept to PRO+ (Production)–also sometimes written as OPS+–which Pete Palmer published in Total Baseball (and in his upcoming Encyclopedia with Gary Gillette).

For example, a player who strikes out 150 times in 600 plate appearances has a strikeout rate of .25. The MLB average was .1658 (or about 99 strikeouts per 600 PA). His normalized strikeout rate is .25/.1658 = 1.50, or about 50% above the league average.

Similarly, the average MLB player hit about 17 (16.8) HR per 600 PA last year. A player who hit 25 HR in 600 PA has a normalized HR rate of about 1.50. Through normalization, we’ve equated eight home runs with 51 strikeouts, for the purposes of their impact on TTO.

Taking the average of the normalized rates of HR, BB, and SO gives us a new look at the TTO leaders:

                                        TTO%                          NAVG
NAME           PA   HR  BB   SO   TTO%  RANK  NHR   NBB   NSO   NAVG  RANK
Bonds_Barry    550  45  148  58   45.6%   4   2.92  3.15  0.64  2.23    1
Thome_Jim      698  47  111  182  48.7%   1   2.40  1.86  1.57  1.95    2
Edmonds_Jim    530  39  77   127  45.9%   3   2.63  1.70  1.45  1.92    3
Giambi_Jason   687  41  129  139  45.0%   5   2.13  2.20  1.22  1.85    4
Dunn_Adam      469  27  74   126  48.4%   2   2.05  1.85  1.62  1.84    5
Sosa_Sammy     589  40  62   143  41.6%   9   2.42  1.23  1.46  1.71    6
Sexson_Richie  718  45  98   151  41.0%   11  2.24  1.60  1.27  1.70    7
Delgado_Carlos 705  42  109  137  40.9%   12  2.13  1.81  1.17  1.70    8
Thomas_Frank   662  42  100  115  38.8%   17  2.26  1.77  1.05  1.69    9
Lopez_Javy     495  43  33   90   33.5%   56  3.10  0.78  1.10  1.66    10

NHR: normalized HR rate
NBB: normalized BB rate
NSO: normalized SO rate
NAVG: average of the three normalized rates

Banished are the pretenders to TTO greatness, like Bellhorn and Jose Hernandez. The list now contains true HR hitters like Jason Giambi, Sammy Sosa, and Richie Sexson, whose power nobody questions. Javy Lopez, who ranked 56th on the raw TTO% list, rises to 10th with the normalized measure, allowing his exceptional HR% (the best in the majors last year) to shine. And Barry Bonds, holder of the single-season records for both homers and walks, stands alone at the top of the list.

But wait a minute. Is Bonds really a *Three* True Outcomes player? Sure, he hits homers and walks a ton, but he’s not really a free swinger. He strikes out at less than two-thirds the rate of the average player; he’s more Ted Williams than Rob Deer. And, while we’re at it, does Javy Lopez, with just 33 walks in nearly 500 plate appearances, really justify his high ranking?

The problem that even normalizing doesn’t solve is that one very strong category can make up for a weaker one. A player doesn’t need all-around excellence, strength in each category, to rate highly. Lopez’s home run rate, homering three times as often as the average player, more than balances his below-average walk rate.

But these are the Three True Outcomes, not Two True Outcomes That Are Good Enough To Overlook The Weak Sister Category. So we’ll require players to excel in all three areas. We can do this by looking at the three component normalized rates (NHR, NBB, NSO) and ranking players by their weakest category. We have already normalized the rates of production, so we can fairly compare their normalized rates in different categories.

                                    TTO%                     NAVG      NLST
NAME         PA   HR  BB  SO   TTO% RANK NHR  NBB   NSO NAVG RANK NLST RANK
A Dunn       469  27  74  126  48.4% 2  2.05 1.85  1.62 1.84  5   1.62  1
J Thome      698  47  111 182  48.7% 1  2.40 1.86  1.57 1.95  2   1.57  2
J Edmonds    530  39  77  127  45.9% 3  2.63 1.70  1.45 1.92  3   1.45  3
J VanderWal  374  14  46  104  43.9% 6  1.34 1.44  1.68 1.48  18  1.34  4
R Sexson     718  45  98  151  41.0% 11 2.24 1.60  1.27 1.70  8   1.27  5
P Burrell    599  21  72  142  39.2% 15 1.25 1.41  1.43 1.36  42  1.25  6
S Sosa       589  40  62  143  41.6% 9  2.42 1.23  1.46 1.71  6   1.23  7
D Lee        643  31  88  131  38.9% 16 1.72 1.60  1.23 1.52  15  1.23  8
C Johnson    413  20  49  84   37.1% 23 1.73 1.39  1.23 1.45  23  1.23  9
G Jenkins    554  28  58  120  37.2% 22 1.80 1.22  1.31 1.44  25  1.22  10
J Giambi     687  41  129 139  45.0% 5  2.13 2.20  1.22 1.85  4   1.22  11
...
C Delgado    705  42  109 137  40.9% 12 2.13 1.81  1.17 1.70  7   1.17  15
...
F Thomas     662  42  100 115  38.8% 17 2.26 1.77  1.05 1.69  9   1.05  36
...
J Lopez      495  43  33  90   33.5% 56 3.10 0.78  1.10 1.66  10  0.78  109
...
B Bonds      550  45  148 58   45.6% 4  2.92 3.15  0.64 2.23  1   0.64  153

NLST = Lowest normalized value out of NHR, NBB, NSO

Four players ranked in the top 10 of our previous list drop out, and in some cases fall very far. Barry Bonds, who headed our previous list, tumbles down to 153rd, reflecting his un-TTO-like strikeout rate. Javy Lopez’s sub-par walk rate forces him not only out of the top 10, but out of the top 100.

And yet, none of the new entrants to the top 10 are an affront to the ideals of the Three True Outcomes. All of them ranked in the top 25 in raw TTO%. But we have culled those who do not fully embrace each of the Outcomes, leaving only the true adherents.

Our third list yields a third different name atop the standings. Adam Dunn, whose strikeout rate 62% above the MLB average was the *lowest* of the three categories, takes top honors as the most exceptional and well-rounded Three True Outcome producer. However, #2 on the list, slightly behind Dunn, and well ahead of the rest of the field, is our old friend Jim Thome. In fact, Thome ranked first or second on every iteration of the list. And, though Dunn edges him in the the “best worst category” ranking, the fact is that the two are remarkably close throughout, having nearly identical raw TTO% rates, and ranking #1/#2 in that category. So close were Dunn and Thome that one additional strikeout from Dunn and one fewer from Thome would have reversed their ranking in raw TTO%.

With two exceptionally qualified candidates, so close in production, and so far from the rest of the field, it seems only fitting to name Jim Thome and Adam Dunn, co-winners of the Three True Outcome crown for 2003.