Methodology

Johnson Power Rankings is a transparent, data-driven rating for ultramarathon runners in the 200+ mile tier (continuous point-to-point/loop races, backyard ultras of ≥48 yards, multi-day/timed efforts of ≥200 miles, and stage races of ≥200 miles total). Rankings publish every Monday. Every parameter below is versioned — current version: v1 (initial parameters (v1)).

How a result is scored

For each ranked result:

score = base_distance_points(class) × difficulty_final(edition) × field_strength(edition) × performance_quality(result)

• difficulty_final — a baseline computed from the course’s GPX (elevation gain/loss per mile, distance, surface type), the race’s historical finish-time spread, and a per-edition weather/heat-cold index — optionally adjusted by a documented expert override (shown publicly on the race’s page). A flat, road, exactly-200-mile course in benign weather ≈ 1.0; the gnarliest mountain/altitude/heat 200s exceed 2.5.
• field_strength — an iterative aggregate of the edition’s participants’ current ratings (so beating a stacked field counts for more than winning a thin one). 1.0 ≈ an average 200+ field.
• performance_quality — for continuous races, your time relative to a course-and-difficulty-adjusted reference; for backyard, a function of yard count; for multi-day/timed, miles covered vs an event-length reference. DNFs score on the fraction of the course completed, damped — non-zero, but never above a comparable finisher.

How an athlete’s rating is built

Your rating is the time-decayed weighted mean of your best 3 performance scores within a rolling 36-month window: full weight for the first 12 months, then a linear taper to zero by month 36. Race rarely? No problem — a quiet stretch doesn’t hurt you until your scoring results age out. New runners start from a conservative prior and earn their way up (no unknown debuts at #1).

What you see

Separate Women’s and Men’s boards; format tabs (Continuous / Backyard / Multi-day-Timed) ranked in their own pools; age-group brackets (20–29 … 70+); and geographic boards (Americas / Europe / Asia / Africa / Oceania). Movement arrows show your change vs the previous publication. Every rank links to the results that currently contribute to it and each result’s base × difficulty × field_strength × quality breakdown.

Current parameters (v1)

best_n

3

window_months

36

decay (start → end)

12 → 36 months

base_distance_points

200+:1000 · 100–199:600 · 50–99:350 · other:200

dnf_damping

0.35

weather_weight

0.35

field_strength iterations / damping

8 / 0.6

cold_start_prior

50

quality clamp

[0.15, 2]

Scope & what’s next

v1 covers the 200+ mile tier. Coming: a curated exception list for historically world-class sub-200 races (Spartathlon, the Hong Kong Four Trails, …), then a drop to 100+ miles, then 50+. Disagree with a rating? Good — argue about it. The methodology is the same for everyone and the math is on the table.