Are retrospective results the same as published picks?
No. A retrospective result is calculated after the event; a published pick has a selection recorded before the event. Both can help describe a model, but they provide different evidence and should not be combined without a clear distinction.
Timing changes what the evidence proves
A retrospective calculation can show how a specified method behaves on historical data. It does not by itself prove that the same selection was available to a reader at the time.
For a pre-event record, inspect the original terms and publication time. A later reconstruction of a price or forecast is not a substitute for that evidence.
Preserve corrections
A correction should identify the original entry, what changed, why, and when. Quietly replacing an old selection makes it harder to understand what a reader actually saw.
A correction can be legitimate and still change the interpretation of a summary. Keep the reason visible rather than presenting only the newest total.
How Tikata labels the difference
Tikata labels results graded retrospectively and keeps them apart from picks recorded before kickoff. The NFL Week 2 results were graded this way. They are shown labelled as graded retrospectively, are never added into the count of picks locked before kickoff, and the two are reported as separate figures. This page carries neither figure; read the running record in the product for what is currently available. Its evidence gate continues measuring; publication is not a declaration that the gate has passed. No historical count on this page is presented as a current result.