Whitey Ballby Dan Fox
March 15, 2006
"The only way to make money as a manager is to win in one place, get fired and hired somewhere else." - Whitey Herzog
On July 25, 1975 Whitey Herzog replaced manager Jack McKeon who had been fired by the Royals after a 50-46 start.
Thus ushered in the era of “Whitey Ball.”
That era would see the Royals finish 41-25 in 1975, good for second place followed by stellar 90-72, 102-60, and 92-70 records and division titles each of the next three seasons.
Whitey was rather unceremoniously let go after the 1979 season and a second place finish, but he quickly moved his style of play down I-70. He won three NL pennants (1982, 1985, and 1987) as both the manager and general manager in St. Louis, where he stayed until midway through the 1990 season. Whitey went on to write a book titled You're Missin' a Great Game, where "The White Rat", as he's sometime's known, extolled the virtues of his favored style of play and took his shots at free agency, the draft, and the Wild Card, among other aspects of the game he felt needed improvement.
In this post-Moneyball age where on-base percentage is king and Earl Weaver, with his veneration of the long ball, is more a folk hero than Herzog, it’s probably appropriate to get in the way back machine and take another look at just what it was about “Whitey Ball” that gave it such cache.
What is Whitey Ball?For those readers not old enough to remember, “Whitey Ball” is often described as consisting of five philosophies: stealing bases, bunting, hitting the ball the opposite way, utilizing the hit-and-run, and relying on fleet-footed defenders. In other words, Whitey Ball of the 1980s was essentially small ball writ large. The Cardinals teams of the 1980s were packed with position players who would ostensibly fit this style of play, including Willie McGee, Tom Herr, Ozzie Smith, David Green, Lonnie Smith, and later Andy Van Slyke, Vince Coleman and Terry Pendleton.
So let’s take a look at those three pennant-winning teams of the 1980s. First, the raw numbers and how these three teams ranked in the National League:
W L RS Rank SB Rank E Rank DER Rank SH Rank HR Rank 1982 92 70 685 5 200 1 124 3 0.710 2 87 4 67 12 1985 101 60 747 1 314 1 108 1 0.718 1 70 7 87 11 1987 95 67 798 2 248 1 116 2 0.696 7 84 2 94 12
All three teams led the league in stolen bases and were near or at the top in fewest errors committed and ranked at least respectably in Defensive Eficiency Ratio (DER). Two out of the three seasons they were dead last in home runs, yet they were near the top in runs scored. From these very basic categories it appears that stolen bases and good defense, in other words team speed, were indeed keys to their attack, while bunting (at least for sacrifices) may not have played as big a role as some remembered. Of course, although not shown in the table all three teams also featured good pitching and ranked third, second, and fifth in ERA respectively. The 1985 team was especially effective on the mound led by the trio of Joaquin Andujar (21-12/3.40), John Tudor (21-8/1.93), and Danny Cox (18-9/2.88),
Overall, the 1985 edition, also the best team of the group and the one I’ll be concentrating on, seems to have epitomized the Whitey Ball approach by stealing a ridiculous 314 bases led by Vince Coleman’s 110, committing just 108 errors, and recording a .718 DER, all of which led the league. All the while the 1985 Redbirds eschewed the home run, hitting just 87 with only Jack Clark (22) reaching 20 and Van Slyke, McGee, and Darrell Porter in double figures, with Porter topping out at 13.
Interestingly, the 1985 team was also the youngest in terms of position players in the NL averaging just 27.6 years.
However, that’s not the end of the story.
Long Live the King?What is often left out of the equation is that all three of these teams, including the 1985 club that led the league in scoring, also led their league in on-base percentage, a point brought to my attention by Dayn Perry's new book Winners.
OBP Rank 1982 0.337 1 1985 0.338 1 1987 0.343 1
The 1985 club did so in a park that was just under league average in scoring (a 99 batter park factor), while the 1982 and 1987 versions were just above at 101 and 102 respectively. And lest you imagine that their speed combined with playing on the carpet at Busch Stadium boosted their run scoring, the 1985 team scored 4.42 runs per game at home, good for second, but a whopping 4.80 runs per game on the road, a full .20 runs better than the second-ranked Astros.
In terms of OBP the 1985 team was led by Jack Clark, who walked 83 times and recorded a .393 OBP while getting help from regulars Willie McGee (.384), Tom Herr (.379), Ozzie Smith (.355) and better-than-league-average contributions from bench players Tito Landrum (.356 in 181 PA), Mike Jorgensen (.375 in 146 PA), Lonnie Smith (.377 in 115 PA), and of course Cesar Cedeno, who hit .434 in 84 PA after being acquired from the Reds on August 29 for minor leaguer Mark Jackson.
So did the Cardinals of the mid-1980s, and particularly 1985, score all those runs because of their speed and small ball approach, or simply because they got more runners on than other teams, or a combination of both?
What About Speed?To look at this question I ran my base running framework for 1985 in order to quantify the contribution of the speed of McGee, Coleman, and company. After all, you'd think that a team that steals 314 bases would also take great advantage of their speed to procure extra bases, thereby scoring runs where their opponents couldn’t.
As a quick reminder, the base running framework takes into consideration plays where there is a runner on first with second base empty and the batter singles, a runner on first with second empty and the batter doubles, and a runner on second with third empty and the batter singles. In each scenario the runner has the opportunity to take one or more extra bases and therefore increase his team's chances of scoring.
For each scenario I then total the number of opportunities for each player and team, calculate how many bases are expected to be gained by the runner given the context of each situation and taking into account the number of outs and which fielder fielded the ball, and then finally convert each opportunity to a number of runs gained or lost based on a Run Expectancy (RE) table that contains the 24 possible base/out situations. For this analysis I used the RE table provided by Pete Palmer and John Thorn in The Hidden Game of Baseball covering the 1961-1977 period.
Unlike the analysis provided for The Hardball Times Baseball Annual 2006, I did not calculate park effects for this article.
The results for the 1985 Cardinals are in the following table, where the ExR column is the number of runs the player would be expected to contribute given his opportunities, BRR is the base runner runs (the total number of runs to which the runner can be attributed given his actual base running performance in his opportunities), and IR is the difference between ExR and BRR and represents the number of runs the player contributed above what was expected.
Opp Bases ExR BRR IR Terry Pendleton 43 76 7.2 11.4 4.2 Willie McGee 56 97 6.4 7.7 1.3 Tom Herr 52 85 9.1 10.3 1.2 Andy Van Slyke 30 45 4.3 5.3 1.0 Vince Coleman 51 73 5.3 5.9 0.6 Tom Lawless 7 11 1.0 1.5 0.5 Lonnie Smith 11 20 1.7 2.1 0.3 Danny Cox 3 6 0.2 0.5 0.3 Randy Hunt 1 2 0.5 0.7 0.2 Ozzie Smith 40 63 9.4 9.5 0.1 Cesar Cedeno 6 8 0.5 0.5 0.1 John Tudor 4 7 0.9 0.9 0.0 Joaquin Andujar 2 2 0.1 0.0 -0.1 Kurt Kepshire 1 1 0.1 0.0 -0.1 Mike LaValliere 1 1 0.1 0.0 -0.1 Ivan DeJesus 7 10 0.9 0.7 -0.3 Steve Braun 5 7 0.4 0.1 -0.3 Mike Jorgensen 10 13 0.8 0.5 -0.3 Brian Harper 3 4 0.3 0.0 -0.3 Bill Campbell 2 2 0.4 0.0 -0.4 Curt Ford 3 3 0.7 0.0 -0.7 Tito Landrum 17 23 2.2 0.8 -1.4 Tom Nieto 14 16 2.0 0.5 -1.6 Jack Clark 48 65 7.8 5.6 -2.2 Darrell Porter 22 31 2.8 0.5 -2.3
In looking at this list you can see that Pendleton, McGee, Herr, Van Slyke, and Coleman as a group contributed an additional 8.3 runs over what would have been expected. Unfortunately, that gain was almost wiped out by Clark, Darrell Porter, Tom Nieto, and Tito Landrum, who combined to “contribute” -7.5 runs.
As a team the Cardinals ranked third in the National League but were a tenth of a run under what would have been expected given their opportunities as shown in the following table.
Opp Bases OA ExR BRR IR HOU 446 706 11 70.6 73.9 3.3 LAN 476 709 14 66.7 69.3 2.5 SLN 439 671 9 65.0 65.0 -0.1 NYN 469 702 21 75.1 72.5 -2.6 MON 383 568 9 54.6 52.0 -2.6 CIN 409 614 11 63.9 60.8 -3.1 PHI 400 574 8 62.6 59.1 -3.5 SDN 468 695 12 72.5 68.8 -3.7 CHN 369 536 7 57.4 51.7 -5.7 ATL 410 604 9 62.1 56.0 -6.1 PIT 403 586 15 66.4 53.3 -13.1
Although surprising, from this analysis it appears that the 1985 Cardinals didn't take advantage of their great speed to advance on the bases more than your average team. From my previous analysis I found that the spread from the best to worst teams in terms of IR is around 30 runs (-15 to+15) or roughly three wins in a single season.
Looking at their stolen bases, however, sheds a different light on the picture.
Following the same type of procedure I calculated the net expected runs from the Cardinals stolen base attempts for each base/out situation using the 1961-1977 RE matrix. For example, a steal of second with nobody out yields on average .285 runs, while getting caught costs the team -0.534 runs. The same steal of second with two outs only increases run scoring by .139 runs while costing the team -.209 runs since the incremental value of moving to third is lessened as is the impact of the third out. The totals for the 1985 Cardinals are found in the following table.
SB Att Run Value Vince Coleman 135 14.4 Willie McGee 72 7.3 Tom Herr 34 5.5 Andy Van Slyke 40 5.1 Ozzie Smith 39 3.7 Lonnie Smith 18 0.9 Cesar Cedeno 6 0.5 Darrell Porter 7 0.3 Joaquin Andujar 4 0.3 Curt Ford 1 0.2 Tom Lawless 3 0.1 Danny Cox 0 0.0 Kurt Kepshire 0 0.0 Mike LaValliere 0 0.0 Steve Braun 0 0.0 Brian Harper 0 0.0 John Tudor 1 -0.2 Bill Campbell 1 -0.2 Ivan DeJesus 4 -0.3 Mike Jorgensen 3 -0.3 Randy Hunt 1 -0.4 Terry Pendleton 29 -1.3 Tom Nieto 2 -1.4 Tito Landrum 5 -1.5 Jack Clark 5 -1.7 Total 410 31.1
Overall, the Cardinals stolen bases contributed an additional 31.1 runs or around three wins.
As it turns out, this accords nicely with the weights Palmer and Thorn assigned to the stolen base and caught stealing when developing their Batting Runs formula. There a stolen base was credited at +.20 and a caught stealing at -.35. Therefore when you apply these weights to the Cardinals' total of 314 stolen bases and 96 caught stealing, you come up with 29.2 runs. One factor that plays into the slight difference is that Coleman attempted stealing third base 34 times on the season and was caught only four times, which yielded an additional 4.3 runs since steals of third with nobody out increase run scoring more than steals of second.
The breakdown of his 14.4 stolen base runs went as follows:
Base SB Run Value CS Run Value Second 78 18.5 20 -8.3 Third 30 6.6 4 -2.3 Home 2 0.5 1 -0.5
And of course, now that we have both the number of runs the Cardinals players contributed with advancing on hits and the amount contributed through stolen bases, we can total the two to come up with a true “speed score” for each player.
IR value Total Vince Coleman 0.6 14.4 15.0 Willie McGee 1.3 7.3 8.6 Tom Herr 1.2 5.5 6.7 Andy Van Slyke 1.0 5.1 6.1 Ozzie Smith 0.1 3.7 3.8 Terry Pendleton 4.2 -1.3 2.9 Lonnie Smith 0.3 0.9 1.2 Cesar Cedeno 0.1 0.5 0.6 Tom Lawless 0.5 0.1 0.6 Danny Cox 0.3 0.0 0.3 Joaquin Andujar -0.1 0.3 0.2 Kurt Kepshire -0.1 0.0 -0.1 Mike LaValliere -0.1 0.0 -0.1 Randy Hunt 0.2 -0.4 -0.2 John Tudor 0.0 -0.2 -0.2 Steve Braun -0.3 0.0 -0.3 Brian Harper -0.3 0.0 -0.3 Curt Ford -0.7 0.2 -0.4 Bill Campbell -0.4 -0.2 -0.6 Ivan DeJesus -0.3 -0.3 -0.6 Mike Jorgensen -0.3 -0.3 -0.6 Darrell Porter -2.3 0.3 -2.0 Tito Landrum -1.4 -1.5 -2.9 Tom Nieto -1.6 -1.4 -2.9 Jack Clark -2.2 -1.7 -3.9
The total is 31 runs, 31.1 from the stolen bases and -.1 from their base running.
As might be expected Coleman contributed almost twice as many runs with his speed and Pendleton, because his poor stolen base percentage (17 stolen bases in 29 attempts) nets only 2.9 runs. Meanwhile Landrum, Nieto, and Clark all sink a little lower because of their combination of poor base running and poor base stealing.
A second issue that one might associate with team speed is the ablity to bunt for hits. Perhaps the Cardinals regularly got runners on via bunt hits before stealing bases to move themselves into scoring position. I took a look at this for the 1985 club and produced the following table showing the number of bunt hits and bunt hit attempts during the season.
BH Att Smith, L 3 9 Smith, O 3 4 Coleman 2 3 Herr 3 3 Nieto 2 3 Van Slyke 2 2 McGee 1 1 Clark 0 1 Andujar 0 1
While the bunts were highly successful (16 of 27), they were hardly a general purpose strategy.
One other consequence of team speed that might come to mind is that having such a collection of speedsters on base would serve to distract pitchers and defenders and thereby afford Cardinals hitters a better chance to drive them in. Studies by our own John Walsh and in the newly published The Book: Playing the Percentages in Baseball all confirm that while there is a definite advantage in opening up the hole between first and second by virtue of forcing the opposing team to hold runners close, that advantage is largely wiped out when the runner on first is a stolen base threat. So it would appear that in general the hitters following base stealers don't derive any significant benefit (see note below).
Final ThoughtsSo what does this tell us if anything about Whitey Ball?
Well, interestingly, even though the Cardinals of 1985 were loaded with speedsters, they didn't seem to take advantage of that speed on the bases or by bunting for hits. However, by stealing over 300 bases they added around 30 runs or three wins to their total. In the big scheme of things their speed, while certainly a component of their league-leading DER, contributed just 4% of their 747 runs. As a result, the speed component of Whitey Ball is certainly overshadowed by the fact that Herzog's teams were populated with a combination of high average and patient hitters. For example, the 1985 club collected 586 walks, some 60 more than average, while leading the league by hitting at .264.
From this it's difficult not to feel that Whitey Ball was more style than substance and that plain old good hitting was the key offensive component to those great Cardinals teams.
References and Resources
In The Book, the authors specifically find that with fewer than two outs and a man on first, the batter gains a 14-point advantage in wOBA (a weighted version of OBA). However, they also found that the advantage disappeared for younger hitters (under 25) and when a "disruptive" runner occupied first base, defined as one with a high rate of stolen base attempts per opportunity. The advantage remained, however, with slow, normal, or fast non-disruptive runners. Interestingly, they also found that stolen base attempts during an at-bat were correlated with a swing in wOBA of a whopping 22 points (14 points from the advantage gained with the runner on first and eight points for a stolen base attempt).
Dan is the author of the blog Dan Agonistes and welcomes your comments and suggestions via email.