ProTriStats methodology
Sources and Freshness
ProTriStats is built around official-source-first data. Each public page should make it clear where the data came from, when it was imported or verified, and whether any important pieces are missing.
- Official race and ranking sources are preferred whenever they are accessible and legally appropriate to use.
- Manual imports are allowed only when they use the same normalized data contract and can be tied back to a named source.
- Missing data is shown as missing data, not as zero, last place, or a silent blank.
- Freshness is page-level, because one race can be current while another series or ranking snapshot is stale.
What counts as a source
A source can be an official race results page, an official ranking page, an organizer PDF, a timing provider page, or a manually prepared import that was checked against one of those sources.
Each imported record can store a source name, source URL, source event id, import timestamp, verification timestamp and quality status. Public pages use those fields for the small data note below result tables and rankings.
Freshness labels
"Imported" means the data was loaded into ProTriStats. "Verified" means it was reviewed against the source after import. When only an update timestamp exists, the page should treat that as a weaker freshness signal.
A race page can be useful even with partial splits, but it should say so. A ranking page is more sensitive: if row counts or source dates are incomplete, the table should show a visible warning or remain hidden.
Manual verified imports
Some official sites are JavaScript-heavy, rate-limited, inconsistent, or unsuitable for scraping. In those cases, a CSV/manual import path is acceptable when it records the same source metadata and import run logs.
Manual import is not a shortcut around trust. It should produce dry-run warnings, unresolved-athlete lists, duplicate checks and item-level failure records before committing data.
- - Historical races may have incomplete split data or no transition times.
- - Official sources sometimes revise results after penalties, appeals, or timing corrections.
- - Different organizers use different labels for elite, pro, DNF, DNS, DSQ and penalties.
- - Show source attribution wherever a user is likely to make a judgment from the data.
- - Prefer explicit "split unavailable" and "data incomplete" labels over polished-looking uncertainty.
- - Keep import logs and data-quality issues available internally so corrections are traceable.