• Book Excerpt: Jaffe’s “Evaluating Baseball’s Managers” — Casey Stengel

    Posted by on December 17th, 2009 · Comments (4)

    Chris Jaffe’s new book “Evaluating Baseball’s Managers – A History and Analysis of Performance in the Major Leagues, 1876–2008” is now available.

    There have been a number of excerpts from Jaffe’s work available online so far – but, we get to feature the one on the Ol’ Perfessor here!

    Many thanks to Chris Jaffe for allowing us to share this excerpt. And, if you like what you see here, then you should get yourself a copy of “Evaluating Baseball’s Managers.”

    Casey Stengel

    W/L Record: 1,905-1,842 (.508)

    Managed:
    - Full Seasons: Brooklyn 1934-36; Boston (NL) 1938-42; New York (AL) 1949-60; New York (NL) 1962-64
    - Majority in: Boston (NL) 1943; New York (NL) 1965
    - Minority of: (none)

    Birnbaum Database: +487 runs
    - Individual Hitters: +224 runs
    - Individual Pitchers: +76 runs
    - Pythagenpat Difference: -21 runs
    - Team Offense: +135 runs
    - Team Defense: +73 runs

    Team Characteristics: Stengel is probably more famous for matching up his pitchers against specific rival teams than any other manager in history. Others actually did it more often, but they did not win five consecutive pennants in the process. He platooned his hitters and used many pinch hitters.

    LPA: 5.33
    LPA+: 120

    Once in a while I encounter someone who thinks managers have no meaningful impact on teams. (This is especially true when dealing with sabermetric types). These people sometimes use Casey Stengel as an example of how overrated managers are. Stengel experienced unprecedented success with the Yankees, but he struggled in his various National League stints as field general. Stengel’s best season in the NL was a mere 77-75 with the 1938 Braves. If Stengel was such a super-genius, then why could he not achieve more when he was away from the Yankees?

    While managers have an impact, the quality of players is much more important. Stengel worked for one great franchise and three utterly dismal squads. The Birnbaum Database gives him a score of +53 runs in Brooklyn and +83 runs with the Braves; rather impressive achievements for such lackluster squads. Stengel had virtually no players to work with in those years. His leading lights were men like Van Mungo and Tony Cuccinello, which is not how clubs win pennants. As a result, Stengel found himself relegated to the Pacific Coast League, where he ran the Oakland Oaks during the 1940s. He had enough success there to land the Yankee job.

    Stengel’s pre-Yankee experiences formed the approach to the game than made him famous in the 1950s. In the NL, he never had any players worth depending on, so he avoided attachments. With Oakland, his most talented players were the ones most likely to be snatched up by the majors, so again he knew not to rely heavily on particular individuals. This outlook became central to how he ran the Yankees. For instance, the 1949 Yankees suffered a litany of injuries with Joe DiMaggio missing half the season, and only Phil Rizzuto playing in over 130 games. Stengel overcame these obstacles to capture the world title by mixing and matching the talent on hand.

    There was no need for Stengel to be too loyal to individual players. If someone declined or could no longer be counted on, he was shown the door. Since the Yankees had a great front office, Stengel could be assured they would land another suitable player to fit his needs. This forced his players to be that much more accountable. Everyone always tries their hardest regardless, but there is nothing like the pressure of imminent punishment to keep people on their toes. As Stengel kept winning, his stature grew so large that his players could not publicly tangle with him. The talent Stengel had on hand was not necessarily better than what Miller Huggins or Joe McCarthy had, yet they never won ten pennants in twelve years as Stengel did.

    Examples of the “what have you done for me lately” attitude abounded on Stengel’s Yankees. Second baseman Billy Martin had as close a relationship to Stengel as anyone on the team, having played for him as a teen in Oakland. In 1956, his career peaked when he earned a selection to that year’s All-Star Game. The following year he got off to a slow start, so the Yankees traded him to the pathologically pathetic Kansas City A’s. New York wanted to break in hot prospect Bobby Richardson and saw no need for patience with the declining veteran. Similarly, after a solid decade under Stengel, Hank Bauer had an off year in 1959. He was in another uniform in 1960. Phil Rizzuto was the anchor for Stengel’s first teams, but once his skills diminished, Stengel would not give him many plate appearances for old-time’s sake.

    Pitching especially demonstrated Stengel’s lack of sentimentality. Tommy Byrne won 30 games for the Yankees in Stengel’s first two years. When he began 1951 poorly, the Yankees sent him to the Browns before the All-Star game. Tom Sturdivant posted two consecutive sixteen-win seasons, but the club shipped him to Kansas City when an injury hampered him. Even Stengel’s most tried and true warhorses were not immune. Among the three hurlers who served as the pitching nucleus for Stengel’s five consecutive pennant winners – Vic Raschi, Ed Lopat, and Allie Reynolds – only Reynolds ended his days in the Bronx. The others were dismissed as soon as they faltered. Stengel’s Yankees had to earn their roster spots everyday.

    With the Mets, this same approach provided the opposite effect for Stengel. Instead of making a great team better, it made a horrible squad worse. He still had no qualms about moving players out of the lineup if they failed to perform. This time, however, he lacked quality replacements. Instead of breeding accountability, Stengel’s technique fostered confusion, which led to apathy. Not only were they not any good, but the players could not get used to their roles. If the Yankees were an ideal situation for his management style, the Mets were the club least suited to it.

    Also, by the 1960s Stengel had lost of a little of his mental edge. The six-month grind of a season demands a certain level of mental alertness and strength that is hard to maintain well into one’s social security years. Only Connie Mack managed at an older age than Stengel, and he was utterly terrible in those seasons. As Mets manager, Stengel scored –504 runs in the Birnbaum Database. Had he retired after 1960, that system would rank him as the fourth best manager of all-time. Instead, he fell to nineteenth place.

    Though continuity existed with how Stengel managed the Yankees and his other clubs, that does not mean he lacked variation in how he handled his different teams. Stengel had engaged in one key strategic habit with the Yankees that was absent in his previous stints: he used a bullpen ace. None of his hurlers recorded more than eight saves in a season with Brooklyn or Boston, but Stengel became baseball’s first manager to preside over ten different occasions in which a reliever logged at least ten saves in a campaign. When the Yankees narrowly won the 1949 pennant, fireman Joe Page earned 27 saves, the most in baseball history until the 1960s. Stengel’s adoption of the fireman model signified a broader change in the game. Though teams had relievers in the 1930s, the concept of a relief ace who specialized in pitching in the most stressful situations did not gain particular attention until Bucky Harris used Joe Page to great effect with the 1947 Yankees. Stengel was one of the first to recognize the importance of what Harris had done, and helped popularize it.

    Stengel treated his relievers the same way he did the rest of his players – with minimal sentimentality. Page helped the Yankees narrowly earn a pennant in 1949, but when he stumbled Stengel swiftly sent him packing. Due to his demanding nature, Stengel became the first manager to oversee three separate relievers log twenty or more saves in a season: Page in 1949, Johnny Sain in 1954, and Ryne Duren in 1958. Until the 1970s, only one other manager (Walter Alston) could make the same claim. Ultimately, eight different pitchers saved at least ten games in a season in Stengel’s dozen campaigns with the Yankees.

    In almost all his career, Stengel leveraged his starting pitchers. He was famous for it with the Yankees, but he engaged in it in his early days as well. In 1934, for instance, Stengel decided the best way to break in rookie Dutch Leonard was by having him face the league’s dogs as often as possible; nine of his 21 starts came against the two worst teams in the league. In Boston and Brooklyn, Stengel was not always consistent the way he used pitchers from year to year. He had a belief in matching starters against particular opponents, but he had no arms he consistently trusted against the best squads. As a result, in 1940 Dick Errickson made eight of his ten starts against the three worst opposing squads for an AOWP+ of 84. Next year he faced the other half of the league for a mark of 106. While working as a swingman for Stengel in 1941-42, Tom Earley posted AOWP+s of 108 and 88.

    Again, the Yankees’ talent fit Stengel’s managerial predilections. The chart below lists what percentage of starts Stengel’s Big Three of Reynolds, Raschi, and Lopez had versus all opponents during New York’s 1949-53 five-peat. As was the case in a similar chart given in the Al Lopez commentary, clubs are ordered from most-to-least wins attained in the period under question:

    Team	Reynolds Raschi	Lopat
    CLE	16.2%	16.3%	20.6%
    BOS	20.0%	18.1%	15.4%
    CWS	10.0%	15.0%	11.0%
    DET	12.3%	11.9%	11.0%
    PHI	10.0%	18.8%	14.0%
    WAS	15.4%	8.1%	14.0%
    STB	17.7%	11.9%	14.0%

    A few interesting patterns emerge. First, Stengel preferred using them against the best available squads. They combined for 75 starts against Cleveland, and 76 versus Boston, but no more than 62 against any other team. Aside from that, Stengel’s usage differed. Reynolds loaded up on starts against the worst two opponents while Raschi almost never faced them. Lopat and Reynolds were more likely to skip turns against the middle class White Sox and Tigers instead of the dregs. Over the entire period, Lopat achieved an AOWP+ of 103, Raschi 102, and Reynolds 100. While none of those scores would be impressive single-season marks, Lopat and Raschi’s totals are rather impressive for a multi-year score. (Even Mordecai Brown ended with a career AOWP+ of “only” 104).

    With the big three gobbling up so many games against the Indians and Red Sox, Stengel could use less trustworthy arms against sad sacks. Tom Morgan was Stengel’s favorite pitcher to deploy against the second division. As a 21-year-old rookie in 1951, Morgan posted an AOWP+ of 92. Next year it was 96. He did not play in 1953, but when he came back the following season, he set a new low of 91. Jim McDonald experienced similar treatment, with back-to-back AOWP+s of 90 and 85 in 1953-54. As a part-time rookie in 1950, Whitey Ford posted an AOWP+ of 89, though by the mid-1950s he emerged as the ace as Raschi, Reynolds, and Lopat went into eclipse.

    In 1954, Stengel’s Yankees, as noted in Chapter 3, achieved the highest single-season LPA by any squad in the twentieth century, 16.52. The chart below shows how many starts each of the seven Yankee pitchers with at least ten starts – Whitey Ford, Eddie Lopat, Harry Byrd, Bob Grim, Allie Reynolds, Tom Morgan, and Jim McDonald – had against each opponent, along with their AOWP and AOWP+ for the season:

    Team	Pct.    WF	EL	HB	BG	AR	TM	JM      Rmnder
    CLE	.721	7	5	0	2	5	1	0	2
    CWS	.610	6	6	4	2	2	1	0	1
    BOS	.448	4	1	4	3	4	3	2	1
    DET	.442	2	1	5	3	1	5	4	1
    WAS	.429	5	4	1	5	0	1	0	6
    BAL	.351	3	3	2	2	5	1	2	4
    PHI	.331	1	3	5	3	1	5	2	3
    AOWP		.533	.518	.439	.459	.508	.431	.403
    AOWP+		112	109	92	97	107	91	85

    (There is no AOWP or AOWP+ for the remainders because that is the total for a collection of pitchers, whereas AOWP-type stats are meant to be assigned to individual hurlers).

    Their record-setting LPA was aided by the league’s extreme winning percentages, but inarguable preferences on matchups abound for every pitcher, except perhaps Bob Grim. Whitey Ford’s AOWP+ of 112 is the best by any pitcher with at least twenty starts in a season since World War II.

    However, Stengel virtually never leveraged his starters with the Mets. As much as Stengel loved to gain that extra edge, the little advantage gained seemed futile with his teams constantly finishing fifteen games out of ninth place. His final fling with leveraging came in 1962 when swingman Craig Anderson posted an AOWP+ of 107 in fourteen starts. Four teams won at least 90 games in the NL that year, and nine of his starts came against them. The Mets went 1-8 in those games, and Anderson allowed five runs in six innings in New York’s sole victory.

    Stengel also had a considerable interest in the double play, both on offense and defense. In his book on managers, Bill James noted that Stengel’s teams constantly turned more double plays than one would expect, even when the middle infielders did not seem particularly impressive. Stengel’s teams also generally avoided hitting into twin-killings. If you compare double plays fielded and recorded for all squads that information exists for, the following teams had the most advantageous single season double play disparity:

    Year	Team	DP	GIDP	Dif.	Manager
    1956 	NYY	214	102	+112	Casey Stengel
    1952 	NYY	199	93	+106	Casey Stengel
    1954 	NYY	198	94	+104	Casey Stengel
    1966 	PIT	215	111	+104	Harry Walker
    1986 	STL	178	83	+95	Whitey Herzog

    The above does not take into account all the adjustments mentioned in Mauch’s commentary. However, even with those refinements, the 1956 Yanks gained the greatest advantage from double plays of any squad in history, and Stengel managed three of the top four squads. (The 1966 Pirates move into second place). Stengel’s teams had the best double play differential in the league eleven times. With the Mauch adjustment, Stengel comes out 537 double plays ahead; easily the largest advantage in baseball history.

    Perhaps Stengel’s dominance in double play differentials is because he had such a long career. Another way of examining double plays is to add together what the Tendencies Database says about GIDP and DP. Based on that, the following managers score best overall with double plays:

    Best at Double Plays – Combined
    Whitey Herzog		1.261
    Billy Southworth	1.357
    Al Lopez		1.415
    Casey Stengel 		1.419
    Gene Mauch		1.453

    The men around Stengel all have a particular hook. Herzog was the king of the base stealers while Southworth and Mauch championed the bunt. Both strategies minimize double plays. Lopez benefited from several years managing one of baseball’s greatest middle infields combinations, Hall of Famers Luis Aparicio and Nellie Fox. Stengel had neither any obvious strategic predilection nor historically brilliant defenders. He just did what he could to make the double play work for him, whether it was positioning his infielders or using groundball pitchers.

    In Stengel’s six years with the Braves, the team averaged over 155 double plays turned per year. In the franchise’s previous 65 seasons, they reached that figure only three times. In his second season, they recorded 178 double plays, a franchise record that stood until the Braves moved to Atlanta. The Yankees averaged 186 double plays a year in Stengel’s dozen years there. In the other 90+ campaigns in franchise history, they turned 186 in a season only two times. In Stengel’s 25 years as manager – and GIDP data exists for all them – none of his batters ever grounded into twenty double plays in a season. He is the only manager who lasted over ten seasons since baseball began recording GIDP information who can make that claim.

    Unlike the double play, Stengel had little interest in the base on balls. According to the Tendencies Database, of all managers who lasted at least a decade, only Burt Shotton’s pitchers scored worse at walks per nine innings than Stengel. In his first year with the Yankees, Tommy Byrne set a franchise record that still stands with 179 walks. The next year he had 160. A few years, later teammate Bob Turley surrendered 177. The five highest single season individual walk totals in Yankee history all came under Stengel, as did one-third of all occasions a Yankee hurler gave up 100 walks.

    Stengel also had tepid marks with offensive walks. Combine his Tendencies Database scores for hitting and pitching walks, and he got as little out of walks as any manager in history:

    Worst with Walks – Combined
    Patsy Donovan 		2.549
    Lou Boudreau		2.472
    Bill Virdon		2.462
    Casey Stengel		2.448
    Bill Rigney		2.434

    Stengel’s teams earned more walks than they surrendered six times in 25 campaigns. Alternately, they had a –100 differential in half-dozen seasons. For his full career, his squads allowed nearly 1,000 more than they gave up.

    Stengel’s score with hitters, 1.061, is technically average, but a lurking variable distorts it: Mickey Mantle. In his decade under Stengel, Mantle constantly appeared among the league leaders in walks, averaging 100 per year in the 1950s. For that reason alone, Stengel should have scored above average. Aside from Mantle, Stengel had only a half-dozen times a batter garnered 80 walks in a campaign.

    Stengel’s walk imbalance partially explained his interest in the double play. Walks allow men to get on base, but are not very good ways to advance runners already on. Stengel would let them have first, provided he could figure out a way to make the next man ground into two outs.

    Comments on Book Excerpt: Jaffe’s “Evaluating Baseball’s Managers” — Casey Stengel

    1. MJ
      December 17th, 2009 | 10:13 am

      Sorry, this all sounds nice but I’m not buying it. Managers are almost entirely a reflection of the players they are managing. They’re capable of losing games but I highly doubt they’re capable of winning them.

    2. December 17th, 2009 | 10:40 am
    3. MJ
      December 17th, 2009 | 11:17 am

      @ Steve Lombardi:
      I read that article. Doesn’t change my mind at all.

    4. Tresh Fan
      December 17th, 2009 | 11:37 am

      Bill James came out with a book evaluating managers some 12 years ago. Two points he made have stuck in my mind:

      1.) The only essential requisite for a manager is to command respect. Everything else is negotiable.

      2.) All managers lose effectiveness over time.

      My own thought is that it seems the basic requirement to be a successful manager is that you were a failure somewhere else.

    Leave a reply

    You must be logged in to post a comment.