How settd works

Canonical Run

We pick the best run from each session to represent your 'setup of the day'. Priority: your pinned run first, then the highest-rated run, then the last run. This canonical run is what feeds into trends, insights, and the Setups list.

Change Tracking

When you make a change between runs and rate the result, we track whether that change helped or hurt. If Run 2 has a higher rating than Run 1 after you changed fork PSI, that's a positive signal. If you changed multiple things at once, each field shares the credit — patterns still emerge over time, but single-change comparisons are the cleanest signal.

Stuck Changes

If you make a change and keep it for 3 or more consecutive sessions without reverting, we flag it as sustained. You've kept it through several sessions, which suggests it's working for you — though sustained behavior isn't the same as proof of effect.

Setup DNA

We look at where your top-rated sessions (4+ stars) fall within your personal range for each setup field. If your best sessions consistently use fork PSI in the bottom third of your range, you tend to run softer. No comparison to other riders — just your own data telling you what works for you.

Anomaly Detection

We flag sessions where a setup parameter is more than 2 standard deviations from your average at that venue AND the rating was low. These are setups that were unusual AND didn't perform well — worth investigating.

Your Data

Your setup data is yours. We never share, sell, or aggregate it. Other users cannot see your geometry, suspension settings, or any other data unless you explicitly share it via a portal link. This is especially important for teams running prototype frames or proprietary suspension tunes.

Geometry Engine

When you adjust headset cups, fork offset, wheel size, or other geometry-affecting parts, our engine cascades the changes through trigonometric calculations: head tube angle affects trail, which affects stability. Seat tube angle shifts with frame tilt. BB height changes with wheel size. Every value shows a 'why' breakdown explaining which adjustments contributed.

Trends

Sparklines on the bike page show your 4 key values (fork PSI, shock, front/rear tyre pressure) across sessions. Change pills highlight the biggest movers. The full trends page shows detailed change cards with old → new values and percentages.

Setup stability

When a setting stays within a small band across your last several sessions, we note it held steady — for tyre pressures within 1.0 psi, fork within 3 psi, shock within 5 psi, over a 5-session window. This is a description of what you kept, not a claim that it is optimal.

Adjacent-run outcomes

On the run timeline, when you change a setting between two runs we describe the next run relative to the one before it — faster or slower (only beyond a small dead-band, and only if you logged times), and higher or lower rated (by at least one star). We never combine these into a score and never claim the change caused the result: warm-up, fatigue, traffic, line choice, conditions, and regression after an unusually good or bad run can all explain it.

Same-context lens

The Insights filter narrows the page to one venue, condition, or rider and re-runs the same calculations on just those sessions. The same sample floors apply to the smaller set, and the denominator always shows how many sessions and runs (and riders) the view is based on. The venue comparison stays on your full history, since comparing venues is its whole point.