Among 115 explicit account links in our selected official-catalog sample, 42 published handles matched the display name after normalization, while 73 did not. This does not make the other links suspicious. It shows why constructing a handle from a person's name is an unreliable way to confirm a destination.
The comparison measures strings, not people. Nicknames, additional words, reversed name order, and numeric handles can all be legitimate when a public source explicitly connects them.
The normalization rule
For both the catalog name and the linked handle, we applied Unicode NFKD normalization, removed combining accent marks, lowercased the result, and removed characters other than ASCII letters and digits. We then compared the resulting strings for exact equality.
This rule removes simple formatting differences such as spaces and punctuation. It does not understand nicknames, initials, transliteration, or identity. The aggregate method record preserves the rule and denominator.
Scroll across the table on smaller screens.
| Comparison | Observations |
|---|---|
| Published links with a corresponding catalog name | 115 |
| Exact equality after the stated normalization | 42 |
| Different strings after normalization | 73 |
The linked sample came from 120 selected biographies, with 60 athletes and 60 comedians. Five had no extracted account link and were excluded from this string comparison, rather than counted as mismatches.
A concrete example of the distinction
The platform biography for Viktor Thorup publishes the surname-first handle thorupviktor. The individual source analysis checks that destination against his public professional presence. Its usefulness comes from the explicit connection, not from rearranging the name and hoping a URL resolves.
A normalized-name guess would preserve first-name-first order and therefore miss this particular string match. The difference is expected once the published handle is observed.
What the result does not measure
We did not test how often fans type guessed names, whether guessed destinations resolve, or whether any differing handle is an impersonation. The 73 differences are not an error rate. They are also not evidence that the accounts changed their handles over time.
Measuring handle changes would require preserved observations of the same account at multiple dates and a reliable continuity check. This study uses one observed catalog sample.
A better verification sequence
Start with the link a credible public source actually publishes. Compare the full destination handle, then check the relevant public identity context. If the source is missing or contradictory, keep the connection unresolved.
The username search guide explains how to use an observed handle. The catalog coverage study explains what an official biography can and cannot establish. Together, these checks reduce the temptation to turn a plausible name into an unsupported account claim.