We Were Wrong About Cleveland.

Out of 1,848 players who changed clubs, the ones who joined Cleveland were more likely to still be playing two years later, not less — ninth of 32. Among established starters, fourth. This is a correction, and the mistake was ours.

The Sports Page · The Professor · marking our own homework

The question: do some franchises actually ruin players, or do they just sign players who were already falling?

The answer: we could not find the effect we predicted. Cleveland sits in the top third for keeping careers going, and does not work its players harder than anyone else works them.

Why that is surprising: we said the opposite, out loud and on the record, and said the numbers would back us.

What you get: the single question that separates a place making people worse from a place attracting people already in decline.

The claim, on the record

Three weeks ago, in a conversation that became notes for this newspaper, one of us said this:

“Cleveland is where players go to die. And they die a horrible death, because they get overused. And I’ll tell you, the stats are going to prove me right about this.”

That was me. It felt obvious. Losing clubs stay losing, the theory went, through impatience and churn, and they burn through people on the way.

Notice that I had committed to a direction before looking. That is the only reason this piece can exist. A prediction made afterwards is not a prediction.

Why the obvious test would have proved me right

Here is the trap, and it would have caught me.

Compare players who arrive in Cleveland against the league, and they look worse. They are worse. Struggling clubs sign players other clubs have finished with, so the men who arrive were already sliding before they got there.

Run that comparison and you confirm the theory every time — whether or not the club does a single thing to anybody. The result is guaranteed by who walks in the door.

So we did it the other way. Take every player who changed clubs, and hold three things fixed: how long he had been in the league, how big his role was the season before he moved, and the year he moved. Then ask a blunt question. Two years on, is he still playing?

What we found

League-wide, 73% of players who change clubs are still in the league two years later.

For Cleveland it is 81% — about six points better than the league once you hold experience and prior role fixed. Ninth of 32. The Jets came 11th. Cincinnati 13th.

Then I tried to rescue the claim. Perhaps the club only ruins good players, and the marginal ones flatter the number. So we restricted it to established starters, men who had played at least half their team’s snaps before moving.

Cleveland went up, to fourth.

The overuse half fails too

The second half of my claim was that Cleveland grinds people down with snaps.

You can measure that. Take a player’s share of his team’s snaps, strip out his own career average, the league’s year, and the ordinary arc of getting older. What is left is the club.

Cleveland comes 15th of 32, a fifth of a percentage point above a man’s own normal workload. That is nothing. The clubs that lean hardest on their players are Jacksonville and Chicago; the lightest are Baltimore and New Orleans, who have the depth not to need anybody for every snap.

Where careers do end

Arizona and Washington. Arrivals there are about 15 and 14 points below the league. New England, Philadelphia and Dallas sit at the other end.

Say it carefully, though. Most of these clubs cannot be told apart — the intervals in the figure are wide, and only the extremes clearly separate. The honest sentence is that Cleveland is not where careers end, rather than that Cleveland is unusually good.

And there is a reading of my own mechanism that turns on me. I said the cause was impatience. Staying in the league is partly a measure of a club sticking with people. By the mechanism I named, Cleveland is one of the more patient organizations in football, and Arizona is not.

What to take home: does the place do it, or attract it?

Whenever somewhere looks like it damages people, ask one question before you believe it.

Is this place making them worse, or taking in people who were already worse?

Hospitals with the worst survival rates often treat the sickest patients. Schools with the poorest results often admit the least prepared children. The manager whose staff keep leaving may have inherited the department nobody wanted. In every case the naive comparison is not slightly wrong; it is wrong in a fixed direction, and it always flatters whoever started with the easy cases.

The fix is the same everywhere, and it is not clever. Follow the same people over time, and hold constant what they looked like before they arrived.

I did not do that before I opened my mouth. I have now, and it says I was wrong.