Not clustered: 21. Jay Kubeil — only 8 of 17 metrics available (needs ≥ 60%); 22. James Atkinson — only 1 of 17 metrics available (needs ≥ 60%); 22. Pete Bolton — only 1 of 17 metrics available (needs ≥ 60%); 22. Adrian Connor — only 2 of 17 metrics available (needs ≥ 60%); 22. Marcus De Vecchi — only 3 of 17 metrics available (needs ≥ 60%); 22. Donny Gardner — only 2 of 17 metrics available (needs ≥ 60%); 22. Dustan Hansen — only 2 of 17 metrics available (needs ≥ 60%); 22. Richard Hughes — only 1 of 17 metrics available (needs ≥ 60%); 22. Scotty Ireland — only 1 of 17 metrics available (needs ≥ 60%); 22. Jason Lannstrom — only 2 of 17 metrics available (needs ≥ 60%); 22. Colin Mackie — only 1 of 17 metrics available (needs ≥ 60%); 22. James McGinty — only 6 of 17 metrics available (needs ≥ 60%); 22. Michael Smith — only 1 of 17 metrics available (needs ≥ 60%); 22. Hossain Tefaili — only 3 of 17 metrics available (needs ≥ 60%); 22. Todd Wisewould — only 4 of 17 metrics available (needs ≥ 60%).
GlideComp groups the pilots by flying style, and not by score. It transforms the rank of every behavioural metric to a percentile inside the field. It then compares two pilots by the mean percentile gap over the metrics that both pilots have, and never fills in a missing value. Ward-linkage agglomeration forms the groups, and the best mean silhouette selects the number of groups. Each group carries the spread of the GAP ranks of its members, which shows where a style paid and where it did not.
k was searched from 2 to 6. The mean silhouette is 0.17 — a value near 0 means soft group boundaries, and a value near 1 means tight, well-separated groups.