I unearthed a bushel of analyses, though conflicting and inconclusive, which all in all suggest that over a lengthy career, an effective manager is worth about one more win per year than the team would otherwise have.

Except for that whole "smell test" things, this might convince. At any rate, "effective" and "otherwise" are typical weasel words.

What's the author's "bushel of analyses"? Actual v. expected wins with a little tuning?

There may be no best manager in baseball, but is there such a thing as the Worst Manager. Ever?

I unearthed a bushel of analyses, though conflicting and inconclusive, which all in all suggest that over a lengthy career, an effective manager is worth about one more win per year than the team would otherwise have.

bs.

I agree the fans theorem of 'He's cost us 8 wins AT LEAST so far this year!!' is insane. But, I have a hard time believing chasing wins (especially when it flys in the face of Tango's work that says 'the 4rth time through the line-up a pitcher should be pulled regardless of the score') and 'the closer -featuring the set-up man!' are only worth 1 win a year. Throw in a liberal sprinkling of bunts, hit and runs and IBB with that too.

I thought Peters would pick a peck of pickled analysis, not a bushel.

There may be no best manager in baseball, but is there such a thing as the Worst Manager. Ever?

The worst manager is a guy who lasted less than a season, not long enough to do too much damage. Someone like Maury Wills.

Thus the person with the worst managerial career - the person who helped cause the most losses over the long haul of his career - is not actually the worst manager, just the guy who lasted long enough to have the worst impact.

Nominees? Connie Mack is a legitimate nominee. He was actually a great manager in his prime, but he hung on 20 years beyond that. At the end, he was falling asleep in the dugout during games and calling on long-retired players to pinch hit.

Jimmie Wilson has the worst record of anyone w/out a Mack-ian prime. John McCloskey finished 200 games under .500 despite managing barely 600 games.

Among recent managers, I'd be sorely tempted to go with Don Baylor. He couldn't handle pitchers, hated kids, and was lax with veterans. He was a self-promoter who was a lousy manager.

Buddy Bell has a pretty lousy track record. I know he's been handled horrible hands everyone he's gone, but he still manages to make the least of them. Random factoids on his managerial career: under him, the Royals became the first non-expansion team since the 1950s to chalk up three straight 100-loss seasons. In his first year, the Tigers allowed 1103 runs - the second most by any team since 1900. With the Royals, Bell had a terrible staff, but the one starter he kept in the rotation all year - Jose Lima - happened to be his worst. That doesn't happen too often.

Maury Wills, followed by Ray Knight

Don Baylor was a surprise for me because he was a clubhouse leader as a player and I think he learned that from Frank Robinson. Both led kangaroo courts . The effin' Red Sox could have probably used a guy like that thi year.

Nominees? Connie Mack is a legitimate nominee. He was actually a great manager in his prime, but he hung on 20 years beyond that. At the end, he was falling asleep in the dugout during games and calling on long-retired players to pinch hit.

You'd think the owner wouldn't have stood still for that . . . .

Ahem . . .

I unearthed a bushel of analyses, though conflicting and inconclusive, which all in all suggest that over a lengthy career, an effective manager is worth about one more win per year than the team would otherwise have.

Except for that whole "smell test" things, this might convince. At any rate, "effective" and "otherwise" are typical weasel words.

And of course, the Braves and Red Sox this year would tell "one more win" can be quite important in its own right.

Among recent managers, I'd be sorely tempted to go with Don Baylor. He couldn't handle pitchers, hated kids, and was lax with veterans. He was a self-promoter who was a lousy manager.

I'm surprised he would grade out that poorly. With the Rockies, he had an expansion team in the playoffs in their third season, and kept them over .500 two more years after that. After Baylor got fired, Jim the Genius Leyland took over, and the Rockies' record dropped by five games.

I don't know anything about Baylor's managerial decision-making (except that one time when he got caught with his pants down in the playoffs), but his record sure doesn't scream "worst ever."

Despite his love affair with Jeff Mathis, Mike Scioscia exceeds his Pythagorean Record every year by a decent margin- usually around 5 games. After 10 years of doing that there needs to be an explanation. Perhaps he is a very good manager.

Just went through Scioscia's record. He's been +1 -2 -2(WS Champions) -3 +1 +2 +5 +4 +12 +5 +1 +1. So, he's been positive for the past 8 years. For his career he's 9/12 and he was below expectations in his one World Series win. He's also +25 over 12 seasons. That's a pretty large sample size to be so far into the positive.

can someone explain to me the theory of why a pythagorean projection should tell us anything about managers, or even a teams ability in close games for that matter? The most obvious problem to me is that what if a team simply wins (or loses) a lot of blowouts (in one direction, w/l/ than the other)? Wouldn't that skew the results and give some team a plethora of runs that they really couldn't parse out and win more games with? There was some article in an old SABR magazine that ranked managers and I guess Yankee skipper Joe McCarthy came out the worse; which might be a case in point.

I started to do some reasearch on upper division teams of both leagues in the 1920s and 30s. Counting blowouts as 8+ run differential, and then comparing this to their record in one-run gamees. Trying to see if like Connie Macks As were really as good as their pythagorean record or whether they had for example a lot of 10 run victories and lost a lot of 1 run games.


But you'd really have to acccount for games that were blown out early vs. ones that were blownout late I guess and I dont program computers or anything. I couldnt really come to any conclusions from the study. I think the Connie Mack As of the early 30s had skewed a lot of blowouts in their favor; but there were other teams that were good and they had about similar number of runs in blowout wins as blowout losses. There was no real trend that I could see, I guess.

In the article he says Bobby Cox might have been worth 2 wins per year. Over 20 years that's 40 wins. I assume this is vs average, not vs replacement level. That's not quite as big as Chipper Jones (55~ wins over average) but still pretty valuable.

Don Baylor has a pass from most Braves' fans. He was supposedly key to Chipper's growth as a hitter, esp switch hitting.

Despite his love affair with Jeff Mathis, Mike Scioscia exceeds his Pythagorean Record every year by a decent margin- usually around 5 games. After 10 years of doing that there needs to be an explanation. Perhaps he is a very good manager.

IIRC, the Angels have been very good base-running teams over the years. Maybe he dials them back in blow outs, resulting in fewer runs. Maybe he's quick to insert the bench warmers/defensive replacements with a lead. Maybe his love of one-run strategies dampens big innings disproportionately more than it hurts W-L record. I have any idea if these are true (I don't watch the Angels), but "beating pythag record = good manager" has always struck me as an assertion supported more by repetition than evidence.

Bill James' wrote something to the effect that most effective managers change the dynamics of an organization (addressing specific pressing problems).

In doing this they change the needs of the organization. And only a handful of managers are able to continually adapt and retain their effectiveness.

Responding to a study that showed most managers were at their most effective in their early years.

#11 The best way to beat pythag (at least the only thing that's proven reliable) is to build good bullpens. I think you'll find that this is a genuine Scioscia skill.

The other thing that might be going on (worth a look) is that most good teams post sensational records in blowouts ( in no small part because good teams don't tend to get blown out )

It's plausible (and happens in strat leagues that have usage limitations) that if you more or less give up on games that are basically hopeless you'll be blown out more often (harmlessly affecting your run differential)

#16 Beating the pythag at least suggests that he's an effective in-game manager. (And that any judgment based on runs dropped has to be tempered by his actual record)

IIRC, the Angels have been very good base-running teams over the years. Maybe he dials them back in blow outs, resulting in fewer runs. Maybe he's quick to insert the bench warmers/defensive replacements with a lead. Maybe his love of one-run strategies dampens big innings disproportionately more than it hurts W-L record. I have any idea if these are true (I don't watch the Angels), but "beating pythag record = good manager" has always struck me as an assertion supported more by repetition than evidence.

Depends on how much and how consistently he beats it. But if he does it nearly every year, and it's by more than just a few games, then either he's a pretty good manager or there's something fundamentally wrong with the whole Pythagorean concept that needs to be adjusted.

I don't put too much stock in beating pythag for evaluating managers because the manager has at least some responsibility for the runs and runs allowed that go into it. He doesn't just work with a fixed set of runs.

As for Scioscia I give up on how he does it. Because if you try and explain it, like strong bullpens, he'll go and use Brian Fuentes + Fernando Rodney for a year and still beat pythag.

Clutch hitting appears to have some association with over and under performing Pythag. As an example, from 2006 through 2010, both the Astros and Angels consistently overperformed their pythagorean record; both teams were also in the top 2 or 3 teams in clutch hitting (as defined by Fangraphs' clutch stat)for each of those years. The Astros' clutch hitting performance dropped to below average in 2011, and the team significantly underperformed its Pythag. Any impact of clutch hitting has to be indirect. My guess is that clutch hitting may lead to over performance in close games and under performance in blow out situations.

Just went through Scioscia's record. He's been +1 -2 -2(WS Champions) -3 +1 +2 +5 +4 +12 +5 +1 +1. So, he's been positive for the past 8 years. For his career he's 9/12 and he was below expectations in his one World Series win. He's also +25 over 12 seasons. That's a pretty large sample size to be so far into the positive.

The +12 jumps out, of course. Sciosa's lucky year, if you will. I'd be interested in a break down of it and see what factors led to it.