you're reading...

A Peek into Managerial Peaks

Players aren’t the only baseball people with peaks to consider. Managers have them too. Because the name of the game with managers is winning titles and ballgames, that peak has, perhaps, even more importance than it does when evaluating their charges. In fact, there’s often more reason to be careful about the context of a manager’s peak.

Think of it this way. We have sophisticated tools that isolate a player from his teammates’ performance and show us what he looks like, himself. But not so with managers. Even things like Pythagorean wins contain from high levels of contextuality. After all, runs scored and allowed by the players form their basis. To what degree are they the manager’s responsibility (and his coaching staff with him) to put the right guys on the field. What degree is that actually the general manager’s responsibility? We use Pythagorean analysis to look at what the manager did with the runs he was given, not whether he pushed his men to score or prevent them. So we do our best with the information we have, and we dive deeper when we can.

How Peak Affects Managers

Take Connie Mack. No one really cares about his last eightteen years, but they are part of his permanent record. You know, the ones where he doddered his way to an 1065–1561 record and a .416 winning percentage. The ones that sunk him under .500 for his career. Wait, what? Connie Mack, the Tall Tactician didn’t even win half his games?! True story. People care about the 30-odd years before that. The decades where he won all of his titles and created two of baseball’s greatest dynasties.

But look inside those first 34 years. It’s a well-worn story that Mack sold off his great early team, and his famous $100,000 Infield. With the rise of the Federal League and the attendant spike in salaries, he couldn’t afford to keep the squad in tact. So he cashed in Eddie Collins and Frank Baker, let Eddie Plank take a more lucrative offer from the Feds and over one offseason went from 99 wins to 43. He slowly, slowly, slowly rebuilt so that by 1925, a mere 11 years later, the Athletics returned to the first division. So here’s what Mack’s career looks like in sections:

EPOCH       TM      YRS       W–L     PCT. TITLES  BEST  WORST  >.500  <.400
Early Days  PIT  1894–1896  149–134  .527    0     6th    7th     3      0
1st Success PHA  1901–1908  639–510  .556    1     1st    6th     7      0
1st Dynasty PHA  1909–1914  583–328  .640    3     1st    3rd     6      0
Interegnum  PHA  1915–1924  528–963  .354    0     5th    8th     0      6
2nd Dynasty PHA  1925–1932  767–452  .629    2     1st    3rd     8      0
Decline     PHA  1933–1950 1065–1561 .406    0     4th    8th     4     10
TOTAL            1894–1950 3731–3948 .486    6     1st    8th    28     16

To be fair, his teams actually got over .500 in three of his last four years, but there’s a lot of speculation that the near-nonagenarian Mack wasn’t really managing those teams except in name. There’s a lot of noise in there. Was Mack just a guy who hung around a lot and caught a few breaks so that he won about a title a decade? No, we can see plainly that Mack’s peak was brilliant and by the number of years it encompasses, if not by consecutiveness, sustained.

In fact, Mack may have the most brilliant peak of them all.

Scaling the Peaks

Using the very same method I outlined in my previous post, I assessed the peaks of nearly 50 top managers. Because managers often have longish careers, I used a broader definition of peak than I do with players:

  • Maximum 10 seasons
  • Had to be a full or damn-near full season (I made a few exceptions)
  • Any season where he won the title or made the championship round
  • Other seasons included on basis of winning percentage to round out the ten years.

Connie Mack’s peak by this method included:

  • a record of 984–515 (5th most wins in the group)
  • a .656 winning percentage (3rd)
  • 235 games over .500 (2nd)
  • 60 wins above Pythagenpat (2nd)
  • a winning percentage 6.5% higher than Pythagenpat would predict (4th)
  • 6 titles (3rd)
  • 8 World Series appearances (3rd).

His closest competitors are the amazing Joe McCarthy and Joe Torre. McCarthy basically never had a bad year, but Torre’s record isn’t so different from Mack’s in some ways. Eleven of his 29 years came with terrible Mets teams or medicore Cardinals squads. But combine his New York and Los Angeles tenures and you’ve got an awfully long and very strong string of excellence. His peak:

  • a record of 984–632 (tied for 5th most wins with Mack)
  • a .609 winning percentage (17th)
  • 176 games over .500 (12th)
  • 56 wins above Pythagenpat (3rd)
  • a winning percentage 6% higher than Pythagenpat would predict (5th)
  • 4 titles (7th)
  • 6 World Series appearances (6th)
  • 10 playoff appearances (tied for 1st)

Torre’s winning percentage is just 17th in the group. But here comes the context. Only four managers who debuted after the 1940s have a higher peak winning percentage: Bobby Cox (.621), Earl Weaver (.615), Sparky Anderson (.615), and Al Lopez (.6093—Torre is .6089). This is an important point for future research by your humble Hall of Miller and Eric electors. How much “easier” was it to be a super team or a super-awful team back in the day? This is potentially a team version of the Schoenfield Effect that we’ve written about in the space before. We’ll talk more about that in my next post, but, yeah, it’s a thing.

The method outlined earlier ranks the managers’ peaks this way:

  1. Connie Mack 18.9
  2. Joe McCarthy 18
  3. Joe Torre 16.7
  4. Casey Stengel 16
  5. Sparky Anderson 15.4
  6. Walter Alston 14.3
  7. Tony La Russa 13.8
  8. Bobby Cox 11.2
  9. John McGraw 10.7
  10. Miller Huggins 10.1
  11. Frank Selee 10
  12. Earl Weaver 7.7
  13. Fred Clarke 7.3
  14. Ned Hanlon 5.2
  15. Cap Anson 4.8
  16. Frank Chance 4.7
  17. Harry Wright 4.6
  18. Tom Lasorda 4.1
  19. Bill McKechnie 4
  20. Dick Williams 3.5
  21. Leo Durocher 2.6
  22. Billy Southworth 2.2
  23. Al Lopez 1.6
  24. Lou Piniella 0.7
  25. Whitey Herzog 0.4
  26. Bill Terry 0.4
  27. Billy Martin 0.0
  28. Ralph Houk -0.1
  29. Danny Murtaugh -0.5
  30. Joe Cronin -0.5
  31. Charlie Comiskey -1.0
  32. Hughie Jennings -1.0
  33. Jim Mutrie -1.5
  34. Mike Hargrove -2.7
  35. Tom Kelly -2.7
  36. Bucky Harris -3.2
  37. Clark Griffith -3.3
  38. Al Dark -3.7
  39. Charlie Grimm -3.8
  40. Red Schoendienst -4.4
  41. Pat Moran -4.4
  42. Steve O’Neill -5.1
  43. Wilber Robinson -5.2
  44. Bill Carrigan -6.5
  45. Jack McKeon -6.6
  46. Cito Gaston -7.3
  47. Herman Franks -7.3
  48. Frank Robinson -8.0
  49. George Stallings -8.7

As we move forward, we’ll be talking about peak more because managers such as Ned Hanlon and Leo Durocher badly need that peak to stay in the hunt. On the other hand, Charlie Comiskey, Jim Mutrie, and Billy Southworth take it on the chin on peak, in part because their either their careers weren’t long enough to encompass ten seasons (Mutrie) or because the meat of their careers (that is, all or nearly all of their full seasons) add to only ten or eleven years anyway, so the good is mixed with the not-so-good. That’s a reality that we will grapple with as we make our decisions. How long was the guy good? Does the excellence outweigh the lack of depth? Are championships more important to us than a long series of pretty good or so-so campaigns that pad up the win totals?

For now, we’re electing the obvious managers in phase one of our managerial search. But we’re glad to tell you about this kind of stuff now so that when phase two comes along, which is one long run-off, you’ll know where we’re coming from.




One thought on “A Peek into Managerial Peaks

  1. I would have never guessed that Selee’s peak would be that much higher than Hanlon, his contemporary.
    Good job, guy.

    Posted by verdun2 | October 7, 2015, 7:58 am

Tell us what you think!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Institutional History

%d bloggers like this: