Scouting guides

Scoutless data and methodology

How Scoutless sources public competition data, calculates alliance statistics, handles missing results, and separates AI observations from recorded evidence.

Public sources and scope

Scoutless imports public competition records through its VEX Events integration, including team and event information, matches, rankings, skills results, and awards where available. The initial public dataset is a current-season snapshot. Profiles reflect the records imported for their displayed context; availability in Scoutless is not proof of complete coverage of an event or season.
VEX Events is an upstream source, not Scoutless. Check the original event record when resolving a score or identity discrepancy. Event recordings can provide separate evidence about robot behavior, but a recording is not a replacement for a published result.

Snapshots and refreshes

Pages use stored competition data. Refreshes import newer source records, and a cached snapshot may remain available when a refresh fails. A page can therefore lag the source; opening or reloading it is not a guarantee that every result was just fetched.
Use a displayed synchronization time or freshness status when assessing recency. The synchronization time describes retrieval, not necessarily the time a match happened or the source last changed. Partial imports, delayed publication, and source corrections can change later results and derived metrics. No fixed live-update interval is promised here.

Final matches and team records

The current dataset service supplies matches with a final lifecycle status to analytics. Both alliance scores must be present. Scheduled, unfinished, and missing-score matches are excluded from score-derived records and alliance ratings. All stored alliance members are currently included; the service does not exclude a member based on its sitting flag. This can limit how accurately a result represents participation.
For a score-derived record, the team must appear in a scored alliance. Its score is compared with the opposing score to count a win, loss, or tie. Win rate is wins / (wins + losses + ties); no usable matches produces null.
The displayed team record is calculated from those stored final scores. Imported official ranking records are shown as separate data and do not override this calculation. They may have different match coverage, so compare the underlying event and match context before treating a difference as an error.

OPR, DPR, and CCWM calculations

For each usable match, the solver constructs one membership row per alliance, with a one for each listed team and a zero for other teams. OPR fits alliance scores to those rows using least squares. DPR fits opposing alliance scores to the same rows. CCWM is OPR minus DPR.
The implementation solves the normal equations with pivoting. If it cannot obtain an independent solution, including a pivot whose absolute value is below 1e-10, it returns null ratings. It does not replace unavailable estimates with zero. The calculation uses the match set supplied to it; it is not an assertion of complete season-wide coverage.
These additive estimates do not directly measure individual scoring, isolate defensive skill, model partner compatibility, or provide a prediction guarantee. The current output has no confidence interval. TrueSkill and Scout Score remain unavailable; no unexplained composite score should be interpreted as a measured result.

AI observations and evidence

AI summaries, video observations, and strategy suggestions are generated interpretations. Keep them distinct from source-reported scores and rankings. A citation lets you inspect evidence; it does not make a generated claim verified or turn it into an official ruling.
Check linked match footage and timestamps before using a claim. Camera obstruction, incomplete footage, incorrect attribution, and limited samples can all affect an observation. Confidence labels are not guarantees. Use “unknown” when the evidence does not show the action clearly.

Corrections and independence

For a suspected error, record the Scoutless page, event and match identifiers, and the source that supports the correction. Contact help@scoutless.app for Scoutless data concerns. Corrections to an upstream record may require contacting its publisher.
Scoutless is an independent service and is not affiliated with or endorsed by VEX Robotics, the Robotics Education & Competition Foundation, VEX Events, or event organizers. Names and marks belong to their respective owners. These guides explain Scoutless and general scouting practices; they do not establish competition rules or official decisions.

Sources and further reading

VEX Events: upstream competition records
Calculation definitions describe the Scoutless analytics implementation. Scouting recommendations are an editorial workflow, and all numerical guide examples are illustrative.
How to scout a VEX teamUnderstanding OPR and alliance statisticsChoosing alliance partners with scouting evidence

Explore the evidence

Browse teamsBrowse events and match resultsExplore the 8110C team profile
The linked profile is a navigation example. Fictional examples in these guides do not describe 8110C.