The sentence ended. The pattern changed.

Detroit won four NFL championships before trading Bobby Layne to Pittsburgh in October 1958. Over the next fifty seasons, it won none and recorded one postseason victory.

There were good teams in that stretch. The Lions reached the playoffs six times in the 1990s, when their median was 8.5 wins a season. Those six appearances produced one postseason win. Any account of Detroit's history has to make room for the teams that could contend and still came away with little to show for it.

The fifty-year declaration attributed to Layne gives that history its familiar frame. The quotation remains unverified: no contemporaneous record located by the Archive securely authenticates or disproves it. The trade and the football results are documented separately.

In 2008, fifty years after the trade, Detroit went 0–16. The Archive reads that finish as an overpayment. The original sentence required the absence of championships; a season without a single win went further.

The sentence expired in 2008. The debt did not.

Four periods in the record

• October 1958 — Origin. The Layne trade supplies the starting point.
• 1958–2007 — The sentence. Fifty seasons served through conventional football failure: weak rosters, front-office turnover, ordinary mistakes, and missed opportunities.
• 2008 — Terminal zeroing. The sentence reaches its end at 0–16, the first winless season of the NFL's sixteen-game era.
• 2009 onward — The aftercurse. Detroit is permitted credible success while being denied closure. Competitive teams and apparent breakthroughs give the pattern something to work with.

That change is the 0–16 Turn. It is the Archive's reading of Detroit's history. It does not explain how a curse operates, and neither of the instruments below produces it as a statistical finding.

One judged. One computed.

Method includes the reporting, source checks, ordinary explanations, and corrections that support the record. Curse Desk also uses two instruments to examine different questions: how difficult an event is to explain, and how long a franchise has gone without results.

Explanation Burden

A bad result can be easy to explain. Detroit's 0–16 season is an example: a bad team lost every game. Its place in this investigation comes from when it happened, not from a complicated explanation of the football. Case 001 scores 1/10 and remains central to the Archive.

Explanation Burden scores explanatory difficulty on two axes:

• Mechanism Complexity: The nonredundant causal dependencies that must be held together to explain the event.
• Closure Deficit: The explanatory resistance remaining after the best ordinary explanation has been applied.

Each axis runs from zero to five. A missed call may have a straightforward explanation. A correctly applied rule that prevents an obvious mistake from being corrected can leave more to account for. The instrument applies to events beyond officiating disputes; a pool report is not a prerequisite.

The scoring rubric is fixed before raters see the events. Descriptions are stripped of identifying details, and the panel includes ordinary events, rejected candidates, and a non-Detroit anomaly as controls. Three independent raters score each axis. We take the median for each axis and add the two medians for a total out of ten, with no weighting.

The score measures neither suffering nor eligibility for the Archive. Annual adjudication decides what enters the permanent record.

The first panel, Pilot A, evaluated thirty events and supplied the fifteen adopted Archive scores. Cases 004, 011, and 018 remain unscored; they were not included in that panel.

Misery Index v3

Detroit's history also needs a league comparison. Other franchises have long droughts, and the same rules have to apply to them.

The Misery Index measures outcome drought across all thirty-two franchises. Its calculations use frozen public records and a fixed specification. Once those rules are set, each team is calculated mechanically. Choosing the rules, weights, and time windows still involves judgment, and those choices are disclosed.

Detroit ranks eighth in the published v3 results through the completed 2025 season. This is a dated comparison, not a live ranking. It provides context for the Detroit investigation; it does not measure the curse or test the 0–16 Turn.

The full publication belongs to Curse Desk at cursedesk.com/misery.

What the pilot supports

In Pilot A, the Archive events had a median Explanation Burden score of 6/10, compared with 3/10 for the pre-2008 events. The permutation test returned p = 0.0099. Within that sample, the Archive events were harder to explain. The result does not establish that they were rarer. The former Rarity Evidence axis was withdrawn after raters failed to interpret it consistently.

There are limits to that finding. The panel covered thirty events. Three descriptions remained recognizable, the raters had prior project exposure, the two groups were sampled differently, and one author who knew the hypothesis wrote the descriptions. Those conditions matter when deciding how much weight to put on the result. The full rubric, scores, and limitations remain on the Explanation Burden page.

What can be checked

The recovered scoring packet, answer key, three complete rating tables, analysis code, seed, and permutation outputs reproduce the fifteen adopted medians and the reported p-value exactly.

That verifies the calculation. It does not reconstruct the entire study. Provider-native responses and session transcripts, exact provider model versions and session identifiers, and server-side packet-delivery and randomization-generation records remain unpreserved.

The Misery Index's preserved research run also reproduces deterministically. Its public audit bundle contains the specification, code, source hashes, tests, and outputs, but it is not a self-contained offline rebuild. Third-party raw snapshots and derived warehouse files are excluded; running the included tests and rebuilding the results requires the complete preserved research tree.

Corrections and withdrawals remain in Errata, alongside the reasons for them.