Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

It Ain’t Over ‘Til It’s Over, Baseball Prospectus’ book on the best pennant races of all time, is available for purchase in stores and also available online through Amazon. If you like what you read here in this sidebar on the chapter covering the 1967 American League’s pennant chase, you’ll love a book with more than 420 more pages of this sort of content, perfect reading for every fan as he or she settles in to enjoy the final stretch drives and then October’s postseason action.

Durocher’s Obsession: Static Versus Dynamic Offenses
by Clay Davenport

When Leo Durocher took the reins of the Giants in 1948, the club was coming off a historic, record-breaking season. In 1947, the Giants had hit 221 home runs, a then-modern record. (Their total was beaten by the Yankees of 1961, the Twins of 1963, the Tigers of 1987, and 35 other teams between 1996 and 2006, which showed how much the game changed in the mid-1990s.) But to the authors of the 1948 Spink Baseball Guide, the home run feat was more noteworthy than the debut of Jackie Robinson (Spink wasn’t exactly out front on baseball’s racial issues).

Durocher considered the home run record a negative, evidence of the club’s devotion to power at the expense of fundamentals. Asked by owner Horace Stoneham to evaluate the team, Durocher’s answer was succinct. “Back up the truck”:

“It ain’t my kind of team,” I told him. We had Johnny Mize, who could hit the ball a mile and couldn’t run. We had Walker Cooper, who could hit the ball out of sight and couldn’t run. We had Willard Marshall and Sid Gordon, who could hit the ball out of the park and could run a little, but not very much.
“Horace, you’re throwing your money away when you pay me. A little boy can manage this team. All you do is make out the lineup and hope you get enough home runs. You can’t steal, they’re too slow. You can’t bunt. You can’t hit and run. I can’t do anything. I can only sit and wait for a home run.”

As soon as he was on board, Durocher promptly started dismantling the team. To a certain extent, he had no choice. Despite their home run record, the 1947 Giants were not a particularly good offense for the number of home runs they hit. Their OBP, the fifth-best in the league and below average, made sure of that. In addition, the roster was old, and changes were sure to come, whether the new manager liked their offensive style or not. Top slugger Mize was 35, and in his last truly productive season. Catcher Cooper was 33 and was falling far short of his 1947 season (35 home runs, a fluke; it was the only time he ever hit more than 20). In replacing some players, the new manager envisioned a team able to do those things he had complained about to Stoneham. Durocher’s team would be faster; they would steal, bunt, and take extra bases, all the little things that establishment baseball has championed and upstarts like Baseball Prospectus have disparaged. In short, Durocher wanted to swap a static offense (wait for the home run) for a dynamic one (call plays, make things happen).

Managers have often preferred the active teams, the dynamic ones, partly because these teams let the manager do something. The only real exception might be Earl Weaver, who preached the doctrine of the three-run homer. The trouble is that you can’t get a really good offense without home runs. It’s like trying to lose weight without exercising, or running for office without raising money. The gain from home runs is so enormous that any other way requires exceptional-and unsustainable-performances. A team would have to hit .400 to get the same offense from batting average alone.

This seems obvious; it takes a lot of singles, doubles, and walks in sequence for a team to score three runs, whereas it takes just two baserunners and a long fly to score three on a home run. But why do so many managers prefer the dynamic style of play? A purist could argue that the natural style of baseball is dynamic, that when baseball became America’s pastime, it was with a fast-paced, swing-away, run-and-catch style of play, not the slow, methodical approach that the homer-and-walk-and-take-pitches crowd advocates. There has been an unfortunate trade-off between the most exciting strategies and the most effective.

The statistics of a team are a reflection of the players, and the statistics of the players are a reflection of their skills. Skills tend to come in certain packages, partly because certain skills naturally go together and partly because skills also determine how the player is taught. In the modern game, stolen bases and triples largely result from a player’s speed (in the dead-ball era, triples were also an expression of power). Triples and speed have a high correlation, but the correlation is far from perfect. Stolen-base king Rickey Henderson was a poor triples hitter, largely because left-handed hitters have a huge advantage when it comes to hitting triples, and Henderson was a righty at the plate. Home run hitters, by contrast, tend to be bigger-the upper body strength enables home runs, but it also means more weight to carry around the field and a slower overall runner. Their best drives don’t go into the gaps or corners; they go over the fence-all of which means fewer triples. Home runs and triples have a negative correlation, meaning that in general the more home runs a hitter produces, the fewer triples he hits.

Fast players, lacking power, are taught to slap at pitches, to put them on the ground and use their legs to beat the throws to first. Power hitters take a ripping swing that sacrifices control for power. Sluggers have a greater need to wait on a cripple pitch (while almost any pitch can be slapped, far fewer can be effectively driven). The hitter isn’t the only one whose strategies change: Pitchers, too, have choices, often throwing much more carefully to players who can reach the upper deck and trying to avoid those cripple pitches the hitters crave. It’s the speed guys who get challenged, because the threat of a single is not enough of a deterrent.

You end up with dramatic differences between the quick guys and the big ones in multiple statistics. It isn’t so much that the fast guys hit lots of singles, but that power hitters get very few. Triples work as a differentiator because the fast guys hit more and the sluggers hit fewer. Both walks and strikeouts line up strongly in the power hitter’s favor. Doubles, however, are not useful indicators; they correlate only weakly with stolen bases and home runs, so they don’t tell us what kind of skill the team relies on most.

We will explore the hidden advantages of the dynamic style of play and why Durocher and so many other managers preferred it. Looking at all the teams, we’ve built a scoring system that relies on the correlations just discussed. Singles, steals, and triples drive your dynamic index up; home runs, walks, and strikeouts drive it down. We’re relying on translated statistics, a system of converting all the values to a common baseline, to take out the differences in the way the game has been played. We’re also limiting ourselves to the years since 1920, because the translations from the true deadball era to the live one are considerably more speculative. Each of the six statistics (singles, steals, triples, homers, walks, and strikeouts) is expressed as a normalized value, meaning that we’ve subtracted the average score and divided by the standard deviation, eliminating the problem of determining how many singles equals a home run. The most dynamic and most static teams, by this method, are shown in Table 10-2.


Table 10-2 All-Time Most Dynamic and Most Static Major League Teams,
Dynamic Index and Other Records
     Most Dynamic
Team            Dyn.Index   W-L   Finish
1992 Milwaukee   20.22     92-70  2nd ALE
1985 St. Louis   19.48    101-61  1st NLE
1975 California  18.82     72-89  6th ALW
1979 Houston     18.78     89-73  2nd NLW
1986 St. Louis   18.73     79-82  3rd NLE
1920 Cincinnati  18.50     82-71  3rd
1944 Washington  17.98     64-90  8th
1943 Chicago(AL) 17.68     82-72  4th
1923 Chicago(AL) 16.72     69-85  7th
1991 St. Louis   16.49     84-78  2nd NLE

     Most Static
Team                    Dyn.Index   W-L   Finish
1927 New York (AL)      -17.10    110-44  1st
1933 Philadelphia (AL)  -16.91     79-72  3rd
1947 New York (NL)      -16.44     81-73  4th
1932 Philadelphia (AL)  -16.24     94-60  2nd
1920 New York (AL)      -16.11     95-59  3rd
1931 Philadelphia (AL)  -15.39    107-45  1st
1992 Detroit            -14.88     75-87  6th
1922 Philadelphia (AL)  -14.85     65-89  7th
2000 Oakland            -14.83     91-70  1st ALW
1999 Oakland            -14.75     87-75  2nd ALW

It is no surprise to see Whitey Herzog‘s Cardinals of the 1980s on this list-the 1986 and 1989 teams rank among the top 25 teams on the extended dynamic list-as they are held up as the epitome of a team that relies on speed instead of power. The Cardinals are the most extreme such team ever assembled, and one of the few that did it successfully. The list-leading 1992 Brewers, managed by Phil Garner, hit just 82 home runs, one of the lowest non-Cardinals totals of the last 25 years, and stole 256 bases, the highest non-Cardinals total of the last 25 years. The team that beat the Brewers in the standings, the Toronto Blue Jays, stole just 129 bases that year, but hit 163 home runs.

The rest of the dynamic teams are not very good; the ten teams averaged 714 translated runs and 83.1 wins (real wins, adjusted only to a common 162-game schedule). Meanwhile, the Giants of Durocher’s ire show up third on the static list, along with the most famous team of all time in first and a host of pennant winners. The static teams average 770 runs, some 56 more than their dynamic counterparts, and 91.2 wins.

A group of ten teams isn’t really enough of a sample, but unfortunately for the speed lovers, in this case it’s a fair sample. The top 100 static teams outscored the 100 dynamic teams by an average of 763 to 711.

So what drives managers to choose speedier teams? What makes up for the drop-off in total offense, according to these managers? If the dynamic style cannot match the totals of the static style, perhaps the speedier teams can be more efficient, squeezing more runs out of their events than the static teams can. This reasoning has some merit: Because the dynamic offenses are faster, they should be better baserunners and produce more runs.

No such trend emerges. Consider the difference between actual and expected runs for the season for the top 10, 50, and 100 static and dynamic teams as shown in Table 10-3.


Table 10-3 Differences Between Actual and Expected Runs
        Top 10  Top 50  Top 100
Static   +7.4    +6.4    +4.2
Dynamic  +7.0    +3.2    +1.4

Indeed, the relationship between speed and runs is quite the opposite; the static teams exceed their expected run totals by more than the dynamic teams did. The flaw in the preceding argument is that stolen bases and better baserunning don’t correlate with more runs, except in faulty memory.

Another argument for dynamic offenses is timing. The most-quoted paeans to little ball speak of always being able to manufacture a run when you need it most, which suggests that these teams can control the timing of their runs to get maximum benefit from them. If this were true, the advantage would show up as a discrepancy between estimated wins, based on runs scored and allowed, and actual wins. In this case, the evidence weakly supports the traditionalists. There is a slight advantage for the dynamic team, in terms of wins minus expected wins (Table 10-4):


Table 10-4 Wins Minus Expected Wins
         Top 10  Top 50  Top 100
Static    -0.70   -0.42   -0.27
Dynamic   +0.60   +0.35   +0.44

For a really good dynamic team, we are looking at about a one-win advantage over a top static team. That is, good offensive timing slightly offsets the roughly 50-run advantage of the static teams. By itself, that certainly wouldn’t be enough of a reason for a team to change its style.

Yet we’ve only been looking at half the game in this analysis. Speed is a vital part of the defensive game; fast teams could make up for their offensive shortcomings by being defensively superior, something that has worked for weak-hitting middle infielders since time immemorial (or at least since Davy Force was a star of the 1870s, entirely for his defense). Does the evidence support the idea that defense gets better with either approach?

Still recognizing that even the best defensive schemes have shortcomings, especially as we go further back in time and away from pitch-by-pitch and even play-by-play records, we see little evidence that dynamic play produces better defense. The top 10 dynamic teams allowed more runs (an average of 702 translated runs), relative to their league, than their static counterparts allowed (673). How can this be so, given what we know about defense and speed?

First of all, we are assuming that the pitchers are equivalent-after all, the selection criteria had nothing to do with pitchers or pitching stats. We may be looking at a second-order result that goes something like this: Since at least the 1920s, teams have recognized the superior value of sluggers for generating runs. Despite the occasional protests of managers, the big bats get the highest salaries and end up playing for richer teams. Speed has always been recognized as a compensation for the inability to slug. Furthermore, richer teams can also afford to get the best pitchers.

The results show this to be the case, since the defensive ratings alone do provide a slight advantage to the fast teams. Let’s look at Table 10-5, which shows the fielding runs above average, translated, for the top 10, 50, and 100 static and dynamic teams. Table 10-6 shows the pitching runs above average, translated.


Table 10-5 Fielding Runs Above Average
        Top 10  Top 50  Top 100
Static    -3.26  -4.80   -3.8
Dynamic   +0.29  +4.32  +12.3

Table 10-6 Pitching Runs Above Average
        Top 10  Top 50  Top 100
Static  +29.8   +31.7   +31.6
Dynamic  -2.1   -25.6   -34.0

An increasing disparity in pitchers is offset, but only partly, by improved fielding; still, the fielding appears to be worth another 10 runs or so in the balance between the two forms (Durocher, as manager of the relatively rich Giants, wouldn’t have had to worry about being able to afford pitching talent). Still, the fielding and timing issues only recoup about half of the difference that was lost in having a speed-based, rather than power-based, offense. So again we have to ask why.

We’ll have to speculate on that one. The question is, how much is a manager worth, with his decisions? More relevant to the argument, how much does a manager think the sum of his decisions over the course of a season is worth? Durocher gave the game away in his speech to Stoneham: “Horace … I can’t do anything.” The one thing a dynamic team provides that a static team does not is opportunities for the manager to feel as if he’s making a difference-he can call for stolen bases, hit and run, and sacrifice.

While it is still up to the players to produce, the manager can easily persuade himself that he made a difference. Given the selective nature of human memory (the times something worked are remembered a lot better than all the times it didn’t), perhaps he feels that he “won” one game a month. That would be six games over a year, a level well above the league-average player. A static team, by contrast, offers few opportunities for in-game management, leading to the oft-maligned “pushbutton” manager, a job, as Durocher said, a little boy could do. Granted, it offers the comfortable out of blaming the players for any shortfalls in performance, but for someone who wants to win-and Durocher was as ferocious as managers ever came on that score-it seems that the (to them) obvious advantages of adding their own skills to those of their players must be a net winner.

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