You’re Outta Here! (Part One)

Bobby Cox was known to get into an argument or two in his day (via jspatchwork).

Bobby Cox was known to get into an argument or two in his day (via jspatchwork).

After a big, splashy announcement followed by months of behind-the-scenes wrangling, Major League Baseball finally put forward its plan for extended instant-replay review for the 2014 season.

The new rules have been hashed over in many places, including at The Hardball Times, and I’m not adding to that chorus today. It was the original proposal, including a manager’s challenge system that has survived in modified form, that got me thinking about the possible tangential consequences of new review rules.

Extended instant replay means managers will have greater recourse against dubious or outright bad umpires’ rulings. There won’t be much point in arguing with an umpire over that call down the foul line if you can just challenge it.

If managers are arguing less with umpires, it follows that umpires will be ejecting managers less often for arguing and showing less forbearance when they do argue. Something that has become part of the game, and the culture, of baseball could be on its way out.

That was my first level of thought on the matter. As usual, baseball is deeper.

The replay rules as recently codified give managers one challenge apiece, with a second if the first succeeds. Also, any play past the sixth inning can be submitted for review at the discretion of the field umpires.

This means there will be plenty of circumstances in which a manager may have to plead with the umpires to review a play, and that pleading may escalate to something the targeted arbiter may not tolerate. And there are always balls and strikes, not reviewable under the new regime and not arguable under the rules.

Also, the manager is not necessarily thinking about approaching the line without crossing it. People get mad sometimes and lose control. Sometimes they get mad with deliberate intent. Plenty of managers say they have intentionally gotten themselves ejected from games in order to fire up their teams in adverse situations. Whether or not they have a challenge in their hip pocket won’t affect that.

So the effect of the replay system on ejections isn’t that easy to forecast. That there will be an effect, or several that combine into one overall pattern, is almost sure. This is why it remains a good time to take stock of the history of manager ejections, since we’re at the cusp of an oncoming shift in that history.

In this article and two more planned to follow, I am going to examine the phenomenon of manager ejections. In this first part, I’ll be giving a more general overview of ejections, revealing who was most and least likely to be run out of a game, and how those answers change when their eras are taken into account.  Later entries will get into deeper detail, looking at how ejections correlate to on-field records, and even whether getting tossed from a game might be good for your team.

The ejecting environment

Ejections have not always been part of professional baseball. It was only in 1889 that umpires were empowered to remove a manager or a player—and back then, a manager was very often a player—from a game, rather than merely fining him for unruly conduct. There were two competing leagues in that era, the National League and the American Association, and they appear to have hit upon the notion around the same time.

The dubious honor of the first manager ejection came on Sept. 15, 1889. Dan Shannon, manager and second baseman for the Louisville Colonels, got himself run out of the first game of a doubleheader in Brooklyn for disputing a play at the plate.  The umpire doing the dis-honors was Fred Goldsmith.

Four days later, in the National League, Ned Hanlon of the Pittsburgh Alleghenys got himself run by umpire Jack McQuaid over a hit-by-pitch call in a game played at Indianapolis.

You may have noticed that both managers were ejected in games on the road. This is little surprise. Umpires in that era endured constant intimidation and even physical threats from players and fans alike. Very few games had more than one umpire presiding, so nobody had his back.

Making calls against the visitors was therefore much safer than doing so against the home team. Indeed, giving their umpires a weapon to combat this intimidation may have been the rationale behind the leagues’ granting them ejection powers.

The following charts show how the frequency of manager ejections per game has changed over the many years since then. First, though, an acknowledgment and a caveat.

Ejection data for this series is taken from Retrosheet. Their coverage of this information is thorough, but it is also ongoing. They have volunteers tracking down the specifics of under-covered ejections, and this occasionally leads to the discovery of unknown ejections or of phantom ejections.

In the few weeks since I originally compiled the data, an April, 1975 ejection of Ralph Houk was found to be spurious. I’ve updated my files, but there probably will be future changes that my series won’t cover.

It’s bad news, because my data set eventually will become obsolete, beyond the accretion of future seasons. It’s good news, because Retrosheet does not rest on its considerable laurels. They’re still working toward the most comprehensive and accurate records of baseball they can produce. I will take my January, 2014 snapshot of that process, and work with it.

Ejec1-RateByYear

If anybody thought managers had grown more sedate since the rough-and-tumble early days of the sport, here’s a strong argument against it. Even after some drops from the all- time peak in 2007, at 2.28 percent, the rate of manager ejection remains historically very high. Whether it’s heading up or down is not clear. The overall trend is upward, but there have been spring tides in the rate before, followed by ebbs.

There are no obvious causes for the big rises and drops. The plunge from the summit of the 1900s-1910s comes after the folding of the Federal League, but there was no corresponding spike when the Feds opened for business.

There’s a big surge in the late ’40s and early ’50s, which continues upward to its peak the year after one of the most often-ejected managers in history, Frankie Frisch, left the dugout. It wasn’t one particular personality driving any of the booms, unless perhaps Earl Weaver was more influential than we knew.

In case the seismic zigzag is tough on your eyes, here’s a further chart consolidating the results into decades.

Ejec1-RateByDecade

Of course, pinning the entire course of this line through the years on managers leaves out half the story. The umpires issuing those ejections had and have plenty of influence on their frequency, possibly more than the managers themselves.

Figuring out how to divide the responsibility would be something like declaring what percent of the game is pitching or hitting: a lot of assertions, with little hard data to back it up. As I’m concentrating on the managers’ side of matters, I’ll talk as though it’s all their doing, but keep in the back of your mind that it’s really not.

Who’s out?

So, which managers have been ejected the most throughout baseball history? It’s a pretty simple question, or at least a simple beginning to analyzing the question more thoroughly. The leader board is below, going to 15 names instead of 10 because this allows me to make an interesting observation later.

Most ejections, baseball history
Rank Manager Total Ejections
1 Bobby Cox 161
2 John McGraw 116
3t Leo Durocher 94
3t Earl Weaver 94
5 Tony LaRussa 87
6t Frankie Frisch 81
6t Paul Richards 81
8 Jim Leyland 72
9 Ron Gardenhire 68
10 Joe Torre 66
11 Lou Piniella 63
12 Clark Griffith 61
13 Bruce Bochy 60
14t Charlie Manuel 52
14t Bill Rigney 52

Cox leads the field by an impressively wide margin, but we’ll cut it down to size soon. Frisch was something of a John McGraw protege, though they had a falling-out over McGraw’s pressure-cooker managing style.

Clark Griffith managed for almost 20 seasons before becoming owner of the Washington Senators. Bill Rigney is somewhat forgotten today; he managed the Giants through their crossing to the West Coast, then helmed the expansion Los Angeles/California Angels for their first eight-plus years.

As you might gather from the charts I posted above this table, raw ejection totals don’t tell the whole tale. One would expect a manager working in the 1930s to get tossed less often than one at the turn of the millennium, just as one expects a modern pitcher to have a higher strikeout rate than one from the Deadball Era. Cox’s huge total stands on a twin foundation of having a very long career and working in a high-ejection period.

So let’s look at this leaderboard through a different lens. I will compare these managers’ ejection rates to the averages for the years in which they managed, pro-rating for partial seasons. The ratio I derive, I will inevitably term Eject+: 100 being average, higher numbers meaning higher rates. Here are the ranks of the well-thumbed again, this time also showing their Eject+ ratings.

Most Ejections, baseball history
Rank Manager Total Ejections Eject+
1 Bobby Cox 161 219
2 John McGraw 116 293
3t Leo Durocher 94 222
3t Earl Weaver 94 290
5 Tony LaRussa 87 105
6t Frankie Frisch 81 374
6t Paul Richards 81 300
8 Jim Leyland 72 124
9 Ron Gardenhire 68 184
10 Joe Torre 66 94
11 Lou Piniella 63 107
12 Clark Griffith 61 202
13 Bruce Bochy 60 110
14t Charlie Manuel 52 154
14t Bill Rigney 52 174

Cox winds up a mere sixth out of 15 on this list in Eject+. It’s still an impressive rate, more than double the league average, but he isn’t looking quite as ferocious as he did before. Weaver and McGraw live up (down?) to their reputations, but Frisch and Paul Richards emerge as the leading hotheads of the group.

Just as instructive is who falls down the rankings. Joe Torre comes in below average, his raw total a pure product of longevity and era. Tony LaRussa similarly hugs the mean, perhaps belying his reputation.

Definitely belying a reputation is the mere 107 for Lou Piniella. Sweet Lou was considered something close to a gold standard for outbursts against the umpires, complete with abuse of equipment like the periodic base toss. Seems he had us fooled, or maybe when he did get ejected, he made it worth the full price of admission.

Does Frisch’s ejection crown stay on when we look beyond the raw number leaders? I’m not about to check every single manager in history, not when there’s partial-season stuff involved that I cannot automate, but I can do a good approximation. I can list the all-time leaders in percentage of games in which they received ejections, an interesting list in itself, then put them through the Eject+ machine.

I went with a floor of 100 games managed to avoid cases like that of Mike Griffin. He managed four games in 1898, got ejected once, and thus, technically holds the record at 25 percent. (I confess myself disappointed that no one-day interim manager got himself heaved to peg the meter at 100 percent.) This semi-arbitrary floor does exclude plausible leaders like Kid Elberfeld (4/98) and Maury Wills (3/83), but we’ll have to live with that.

Along with games, ejections, percentage, and Eject+, I noted the managers’ winning percentages, as a fresh hook for further study. The managers are ranked by ejection percentage.

Highest Ejection Percentage, baseball history
Rank Manager Games Ejections Ejection % Eject+ Win %
1 Bill Dahlen 615 32 5.20 343 .414
2 Larry Schlafly 196 10 5.10 335 .484
3 Burleigh Grimes 306 15 4.90 513 .434
4t John McLaren 164 8 4.88 230 .457
5 Cal Ripken Sr. 169 8 4.73 279 .402
6 Paul Richards 1,837 81 4.41 300 .506
7 Lee Magee 118 5 4.24 272 .453
8 Otto Knabe 314 12 3.82 249 .425
9 Earl Weaver 2,541 94 3.70 290 .583
10 Frankie Frisch 2,246 81 3.61 374 .514

Bill Dahlen was another ex-player for McGraw who also ended up hating him and also ended up managing with the same angry streak. Schlafly, Magee, and Knabe all managed solely in the Federal League, a notable concentration of temper.

All the managers on this list more than doubled their era’s ejection rate, but Frisch’s Eject+ stands up to most of them. However, it cannot stand up to Burleigh Grimes.

Grimes is noteworthy as the last legal spitball pitcher, reaching Cooperstown on a wave of moisture and fisticuffs. He managed Brooklyn in 1937 and ’38, a rather moderate era for ejections, and carried the combativeness he’d shown as a player into his new role. Given the environment, he stands head and shoulders above everyone else as the manager most liable to be thrown out of a game.

After two tempestuous (for the umpires) years of losing baseball under Grimes, the Dodgers had to be looking for someone more easy-going. They hired Leo Durocher. Only in this context could Durocher be considered a calming presence: he followed a man we can call the most ejection-prone manager of all time.

Looking at the winning percentages of these oft-tossed managers, there are distinct patterns. Seven of the 10 had losing career records; the excluded Elberfeld and Wills had winning percentages of .276 and .317, respectively. The three who were winners (two only mildly so) had easily the longest tenures. It’s no surprise that winning managers last longer.

The preponderance of losing managers on the list seems to point to something, but there’s less than meets the eye, and it relates to the previous point. The ratio of losing to winning managers in baseball history is about two to one. (Exactly 452 to 233, to date.)

Losing managers tend to get fired and stay fired, so there are more of them cluttering the minus side of the ledger. The apparent evidence that hothead managers are losing managers doesn’t hold up, at least not with a single top-ten list.

(It’s fair to note here that managers who last at least 100 games, my cutoff point, do slightly better than .500: .50147 to be precise. This doesn’t affect the conclusion. It’s still three managers above and seven below, nine counting Kid and Maury.)

Who’s in?

We’ve concentrated so far on the top of the scale, but there’s also a bottom. Some managers never tangled with the men in blue, or at least knew how far was too far. The lowest number of ejections for a manager is obviously zero, so we have to judge instead on how many games someone managed without getting the heave-ho.

This requires a different kind of cut-off. Managers whose careers began before 1889, when ejection was first permitted, are excluded from the following list. Otherwise, the leader would be Harry Wright of the original Cincinnati Red Stockings. He had 2209 spotless games, but in only five of his 23 years of managing could he have been ejected.

Within that limit, these are the all-time leaders, the managers who worked 500 or more games without once being ejected. Their years of managing, total games, and winning percentages are included.

Zero Ejections, baseball history
Manager Years Managed Games Managed Winning %
Frank Selee 1890-1905 2,180 .598
Pinky Higgins 1955-1962 1,119 .502
Al Buckenberger 1889-1904 1,043 .475
Walter Johnson 1929-1935 966 .550
Bill Shettsline 1898-1902 677 .548
Cookie Lavagetto 1957-1961 657 .412
Ted Williams 1969-1972 637 .429
Mickey Cochrane 1934-1938 600 .582
Dave Foutz 1893-1896 532 .507

To that list, one can add a notable near-miss. On Sept. 6, 1895, in his first full season as player-manager of the Pittsburgh Pirates, Connie Mack got thumbed out of a game in New York by Hall of Fame umpire Hank O’Day over a play at second base. (Yes, another road game.) Mack would manage 7610 more games, 7460 as owner-manager of the Philadelphia A’s, and never be ejected again.

I quote directly from The New York Times of the following day, in case nobody can believe the strait-laced and reserved Mack capable of this singular outburst:

It was a contest that was more or less interesting from start to finish, and at the close of the fifth inning the argument grew so very interesting that the umpire, Hank O’Day found it necessary to fine Manager Mack of the Pittsburgh Club $100 and order him off the field for using insulting and abusive language.

Mack did not feel inclined to go, but O’Day was determined. He called several police officers, who soon convinced Mack that resistance was useless, and he retired from the field amid the jeers of the spectators.

Connie Mack, using abusive language? Connie Mack, having to be escorted off the field by the police? Chalk it up to the follies of youth.

Back on the main list, Frank Selee was the architect of the Boston Beaneaters dynasties of the 1890s, and he then laid the foundation of the mighty Cubs of the 1900s before tuberculosis took him away from the game.

Seeing the gentlemanly Walter Johnson on this list is no surprise, except for those who forgot that he managed for several years. Ted Williams is a surprise to anybody who recalls his clashes with reporters and fans. I suppose if you weren’t booing or taking notes, he had no gripe with you.

There’s no point in giving Eject+ numbers for those nine, as they’d all be zero. Instead, we can calculate the probability of a manager with league-average liability of being ejected getting through all of those games without being asked to leave.  Showing how unlikely it was an average manager would avoid the thumb will reveal how beyond the average these managers were.

(This involves a lot of exponents, and I won’t bore anybody with the formula, but it’s high-school math at worst, reasonably easy to noodle out for oneself.)

The only two real contenders for the top spot are Selee and Pinky Higgins.  The rest just had too few games in eras with too few ejections. For Selee, the chance of an average manager working his games without suffering an ejection is a minuscule 0.00346 percent, or if you please, about 28,911-1 against. That’s amazing, yet it is nowhere near Higgins’ mark. Working in a much more ejection-prone period, he comes in at a sub-atomic 0.0000249 percent, or just above 4,000,000-1.

Higgins had a 14-season career as a player before managing the Boston Red Sox for eight years. He participated in 2921 major league baseball games without once being ejected. His first five years were with the Philadelphia A’s, so he may have learned something from Mack, once Mack had settled down from his wild youth.

Those eyeing the numbers may have noticed that six of the nine top un-ejected managers had winning records. Again, this seems to point somewhere, and again it doesn’t do so as much as it appears.

These managers are all above a 500-game cutoff, and managers with at least 500 games have an overall .510 winning percentage. Higgins and Dave Foutz, while being above .500, fall below .510, so it’s only four of nine managers who finish above the mean. Throw in Mack, whose long dry spells drag his record down to .486, and it’s four of 10. Remember that twice as many managers are losers as winners, and this now looks quite unremarkable.

In the future

That’s twice I have tantalized you with what looks at first glance like evidence that ejection rates correlate to one’s success as a manager, one way or another. In the next installment of this series, I will quit teasing and get you honest answers. Are good or bad managers tossed more often? Do individual managers get ejected more often in down years than up years? I will answer those questions and a couple of others.

Further down the line, I’ll be sifting through the internals of ejection games themselves. How well or badly are teams doing in games in which their skipper gets the heave-ho? And maybe most intriguing of all, does a manager getting himself tossed help or hurt his team?

References and Resources

Manager data was drawn both from Baseball-Reference and Retrosheet. Also helpful was the book Evaluating Baseball’s Managers by our own Chris Jaffe. The New York Times of Sept. 7, 1895 provided insight to a unique day in Connie Mack’s career.

Print Friendly
 Share on Facebook1Tweet about this on Twitter7Share on Google+2Share on Reddit0Email this to someone
« Previous: Redefining Batted Balls to Predict BABIP
Next: Lessons in League Switching »

Comments

  1. said...

    Seems like a lot of ejections are more or less staged with the manager purposely getting tossed to make a point to his own players or else to deflect attention away from a player who is on the verge of getting tossed himself.

  2. bucdaddy said...

    While I admit one of the more entertaining things I’ve ever seen in person at a ballpark was an Earl Weaver tantrum over a balk call, I have to say that I never, ever go to a game thinking to myself, “It sure would be cool to see a manager get run today. I really hope one of them holds up the game for 10 minutes to argue a call that has no chance of getting overturned.”

    “Something that has become part of the game, and the culture, of baseball could be on its way out.”

    I really don’t think we’ll miss it one bit. In no other sport I can think of are coaches routinely allowed to charge onto the field, pitch or court and berate the officials at length, and it doesn’t hurt those sports at all. Baseball will get over it.

    In any case, interesting work, Shane. Looking forward to parts 2 and 3.

  3. Carl said...

    I was surprised not to see Billy Martin who, like Sweet Lou Pinella, also had some major ejections and threw various items around the infield.

    Ted Williams was seldom ejected as he was so widely revered by the men in blue.

  4. Paul G. said...

    I’m not totally surprised that the Federal League had some issues with ejections. During some newspaper research I found a couple of articles about the umpiring quality of the FL and, according to the writers, it was quite poor. One incident that stuck with me is an incident where the batter threw his bat at the pitcher and charged the mound, yet no one was ejected. Yes, that’s the opposite extreme, but I suspect there was good reason for managers to get ornery.

  5. John Fain said...

    For at least part of the time Durocher managed, the manager could leave the dugout to argue a ball or strike call. While this would be an automatic ejection now it was just a needless delay of the game. It is surprising how relatively short games were back then with managers popping out of the dugout to argue a ball/strike call and making frequent visits to the mound. Part of it was due to the batters not stepping out of the box after every pitch for a practice swing or two.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *