you're reading...
Managers, Sidebars

It Only Gets Harder for Managers to Win

Last time I wrote about managers, I noted that the Schoenfield Effect is in play with them too. See, it turns out that it’s getting harder and harder to be the manager of a very high—or very low—performing team. The following chart tells the story awfully well.

Screen Shot 2015-10-06 at 8.11.20 PM

This kind of trend is a statheads’ dream in its clarity. As you can see, over the past 150 years, the best and worst teams in the league have gotten worse and better respectively. What’s more, the degree to which they’ve gotten worse and better is on the order of more than 100 points of winning percentage. That’s sixteen wins over a 162 game season, or the difference between the historically awful 1962 Mets and the 2011 Astros who were maligned for tanking en route to their resurgence in 2015. If we look at it closely, we can split the top teams into five distinct epochs.

  • 1871–1885: These teams are crazy good, but the schedule is under 100 games.
  • 1885–1910: The best teams stabilize between .650 and .750, mostly near .700 as the schedule gradually widens to 154 contests.
  • 1910–1926: A brief period of relative parity very close to our own with no team above .700 and most below .650.
  • 1927–Integration: While the best teams did better, they were in an extremely narrow window of about 50 percentage points.
  • Integration–now: Despite small spikes recently, the best teams have stayed mostly under .700.

It’s clear that run environment has a strong impact here because when big changes happen in it, we see spikes.

For the crummy teams we see a pattern of what the QA engineers like to call continuous improvement. Until very, very recently, over most any given period of time, with a few exceptions, the worst team in MLB has had a better winning percentage than its counterpart ten years earlier.

Why? Because baseball, like most sports, is a copycat environment. Once a team has success with an innovation or a strategy, everyone else will adopt it over time. Back in the 1880s when things like practicing, coaching, team play, and basic strategy were still up for grabs, Charlie Comiskey’s St. Louis teams could pull ahead of the pack quickly. But by the 1890s, his Cincinnati teams fell into mediocrity as teams all began to employ his innovations, the bad teams caught up to the good ones more quickly than the good ones could come up with further innovations. This is how today’s Pirates went from losers to winners. They weren’t the first team to adopt analytics, but once they did, they caught up fast. And now every team in the league has analytics, so it’s harder to gain advantages with them. Well, maybe not the Ruben Amaro Phillies.

Now, just to be sure we’re not only dealing with really crazy outliers, here’s a similar chart that shows the average winning percentage of the best and worst teams in MLB each year. The number of teams varies with the number of teams in the league but is always equivalent to roughly the best three and worst three squads in a 16-team league.

Screen Shot 2015-10-06 at 8.23.25 PM

Same exact thing.

We might as well look at one more version of this. Here’s the percentage in MLB of Super Teams and of Super Bad Teams across time. Super Teams had 100 wins in today’s schedule or a .617 percentage. Super Bad Teams won fewer than 60 games or had a percentage lower than .370.

Screen Shot 2015-10-06 at 8.30.36 PM

Screen Shot 2015-10-06 at 8.35.42 PM

More of the same, but charts are fun.

So it’s harder to win or lose crazy numbers of games in a season anymore. This has a subtle and observable effect on how we compare one manger to another across history. So in my next two articles, I’m going to look at some reasons why this phenomenon occurs and then how we might adjust traditional manager statistics to help us look across the decades.



One thought on “It Only Gets Harder for Managers to Win

  1. Part of the reason for really good and really awful teams in the 19th Century is the establishment of Syndicate Ball (one guy owned 2 teams) which allowed for stacking one team with good players and leaving the other to flounder (see the 1899 Cleveland Spiders as the best example).

    And I think a further problem was the early creation of one year or 2 year teams (especially in the American Association) when “pro” teams were popping up everywhere as some guy with money sponsored a team that grabbed a real player or two and a few regional stars and tried to compete with real teams. I know I’m oversimplifying that some because leagues encouraged a certain amount of it. That led to some terrible win/loss percentages. And 3 leagues in 1884 and 1890 didn’t help.

    Having said that, I agree with your conclusions. And you’re right, charts are fun.

    Posted by verdun2 | October 14, 2015, 7:47 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: