Climber analytics: what this page shows

The page /analytics/users/{id} is a deep dive into a single account: four ratings, portfolio structure, history of changes, and terrain profile. It complements the brief version at /users/{id}/ratings with detailed breakdowns and a time-series graph.

If you want to first understand what these four ratings actually measure and when they become official, read the separate guide «How climber ratings work».

The four rating cards

At the top of the page there are four cards: Mastery, Explorer, Form, Altitude. Each shows the same set of fields.

What's in a card

The big number — the rating's raw value. It's a dimensionless quantity: the higher, the higher your place on the corresponding leaderboard. Comparing absolute values across ratings is meaningless — each rating has its own scale.
Rank and percentile — your place in the overall list and the share of climbers you are ahead of. Rank 1 is the very top. Percentile 95% means «top 5%».
Progress bar — graphical representation of the percentile.
«Provisional» badge — appears when you don't yet have enough variety of ascents for the rating to become «official» and put you on the global leaderboard. The number is already computed but doesn't yet participate in comparisons. Details — in the general ratings guide.
Δ over 90 days / Δ over 365 days — change of raw over the last 90 days (for Mastery and Form) or 365 days (for Explorer and Altitude). Positive delta — growth, negative — decline (for example, old strong ascents «aged out» of the 24-month window of Form). null means «not enough history to compare».

Explorer breakdown

Below the main card — four sub-components that make up Explorer:

Routes — how many distinct routes are in your portfolio.
Peaks — how many distinct summits.
Local — how many separate «local» areas (under the hood — H3 cells at resolution 7, edge ≈5 km; in practice this is the level of a single gorge or massif).
Region — how many large mountain regions (H3 at resolution 5, edge ≈45 km; the level of «Caucasus», «Tien Shan», «Pamir»).

The weighted sum of these four contributions yields Explorer's raw. Each sub-component is its own value, not a share; read it as «depth along this axis».

Altitude buckets

Below the «Altitude» card — counters of summited peaks at altitude thresholds:

Highest — elevation of the highest peak you've climbed, in metres.
2500+ / 3000+ / 4000+ / 5000+ / 6000+ / 7000+ / 8000+ — how many distinct peaks you've climbed in each category. Peaks aren't duplicated across categories: e.g., 5200 m falls into 2500+, 3000+, 4000+ and 5000+ — all four counters increase by one.

These counters are computed across all successful ascents, including routes without a confident technical grade. Therefore the count of 4000+ peaks may be greater than what feeds into the «Altitude» rating itself (only routes with a confident grade go there).

Profile structure

The «Profile structure» card has 14 counters describing your portfolio in terms of diversity, activity, and style. All counters are taken from successful ascents (result = full), excluding soft-deleted records.

Diversity

Distinct routes — how many unique routes are in your history. Repeats are not duplicated.
Distinct peaks — how many unique summits have been reached. Counted from the route's «top-3 peaks» (long traverses don't inflate the counter).
Local areas — H3 cells at resolution 7 (edge ≈5 km). One area ≈ one gorge or massif.
Major regions — H3 cells at resolution 5 (edge ≈45 km). One region ≈ a major mountain system.

Altitude and activity

Max altitude — highest peak reached, in metres. — if the climbed peaks have no elevation data.
Total ascents — all successful ascents of all time.
Active months (12m) — how many distinct months in the last 12 contain at least one successful ascent. A «regularity» metric, not a «volume» one.
Last ascent — date of the most recent successful ascent.

Style and role

Lead share — what fraction of ascents went «with your hands» on at least part of the technical sections (you led the crux, or led several different sections, or led all of them). — if no technical role is recorded in the ascents.
Full lead — what fraction of ascents was led on all technical sections. A subset of the previous metric.
Alpine style — share of ascents done in a single push without pre-fixed ropes or intermediate camps. — if style is not recorded in the ascents.
No fixed ropes — share of ascents where no fixed ropes were used (fixed_rope_usage = none). — if not recorded in the ascents.

Repetition and diversity

Repeat ratio — formula 1 − distinct_routes / total_ascents. A value of 0% means «every ascent is a new route»; a high value means «I often climb the same routes». This is a neutral metric — neither better nor worse, just a description of style.
Terrain diversity — Shannon entropy across the terrain types of routes (rock / ice / mixed / snow / combined), normalised to [0, 1]. 0 = strict specialisation in one type, 1 = uniform distribution across all observed types. — if no route has a classified terrain type.

Rating history

The graph shows a time series of raw values for all four ratings. Horizontal axis — months, vertical — raw. Below the graph — a horizon switcher: last 12, 24, 60 months or all time. The horizon choice is saved in the URL, so you can share a link.

How to read the graph:

Line going up = in this period you added significant ascents that raised the corresponding rating.
Plateau = no new ascents that influence this rating, or the influence was offset by loss of weight from older ones (relevant for Form).
Line going down = rare but possible: e.g., for Form, old strong ascents «aged out» of the 24-month window.

Form is the only rating with active «decay»: in the absence of new strong ascents, its line slowly drifts down. The other three ratings are monotonically non-decreasing.

Terrain profile

Distribution of successful ascents by route terrain type:

Rock — rock routes.
Ice — ice (waterfall ice or glacier ice).
Mixed — combinations of rock and ice on the same section.
Snow — snow routes (without significant technical rock or ice).
Combined — routes with multiple distinct terrain types classified.

At the bottom of the card — two indices:

Diversity index (Shannon entropy in [0, 1]) — same value as «Terrain diversity» in the structure card. The closer to 1, the more uniform the distribution.
Specialisation index (max(p_f) in [0, 1]) — share of the dominant terrain type. 1.0 = «100% routes of one type», ≈0.2 = uniform across 5 types. These two indices are opposite sides of the same coin: high diversity → low specialisation, and vice versa.

Data snapshot

At the bottom of the page — a small badge «Computed at …». Analytics is not updated in real time: the recomputation runs once a day, usually overnight. If you logged an ascent today, new numbers appear tomorrow. If the badge shows an old date — the recomputation has been delayed or failed for some reason; try later.

How ratings work — general guide to the four ratings: what they measure, what raises / lowers them, when they become official.
Climber catalogue — the catalogue at /analytics/users with presets, filters, and sorting by the same metrics.
[Climber profile](/en/users/{your id}) — public profile, brief version of ratings and history.
Cohorts and comparison — for regular slices and comparing several accounts.

Frequently asked questions

Why are «4000+ peaks» in my profile higher than the number of peaks in my Altitude rating? The high-altitude counter in the structure section counts all successful ascents, including routes for which we don't yet have a confident technical grade. The Altitude rating itself only includes routes with a confident grade — this is done so that climber comparisons are correct. Details — in the ratings guide.

I have «—» instead of a lead share. Why? The metric is computed only over ascents where you specified a technical role (lead all / lead some / lead crux / did_not_lead). If no ascent has a role specified, there's nothing to divide by — we show a dash. To make the metric appear, fill in the «Technical role» field when logging an ascent.

The 90-day delta shows 0 even though I climbed a new route. Why? The delta is the difference between the raw value now and 90 days ago. If the new route didn't affect this particular rating (e.g., it was already in the portfolio and didn't shift the percentile shelf of Mastery), the delta may stay near zero. It's also possible that the new route compensated for the «aging out» of an older ascent from Form's 24-month window.

How often is it recomputed? Once a day, over your entire history. This applies to all four ratings, all counters, the graph history, and the terrain distribution.

Can I share a link to a specific graph horizon? Yes: the horizon is stored in the URL (?horizon=24); copy the address bar — your recipient will see the same view.