> ## Documentation Index > Fetch the complete documentation index at: https://playbook.smedleygroup.com/llms.txt > Use this file to discover all available pages before exploring further. # Stop Sensitivity > Predicted total race time against stop lap, for drivers picked inside the widget, with the rank-1 strategy's optimum marked. ## Overview Stop Sensitivity plots how much a driver's predicted total race time would change if a given stop happened on a different lap. Pick one or more drivers from the badge strip inside the widget to see one curve per (driver, stop index), with the driver's own rank-1 predicted time marked as the optimum on its curve. > [!NOTE] > > The curve minimum should equal the driver's rank-1 predicted time, within 0.001s, per the engine contract. The optimum marker doubles as a visible consistency check on that guarantee. *** ## What It Shows ![Stop Sensitivity with its driver badge strip on top and the curves for the two default drivers below (Singapore 2026 Race, lap 7)](https://mintcdn.com/smedley_group/1sk8-0h-JRH4PrIf/images/watchparty/stop-sensitivity.png?fit=max\&auto=format\&n=1sk8-0h-JRH4PrIf\&q=85\&s=f1ee1db354df312cba954d68e5776993) The x-axis is stop lap, the y-axis is predicted total race time in seconds. Each series is named ` stop `, one line per stop in the driver's rank-1 strategy, plotted with visible point markers. A dashed horizontal `optimum` mark line appears on a driver's curve at their rank-1 predicted time, when that strategy is reachable. **Choosing drivers.** A strip of driver badges (the three-letter code on the team colour, the same badge the [Strategy Table](/software/internal/watchparty/strategy/table) uses) runs along the top of the widget, one for every driver in the strategy snapshot, so it works before the race starts too. Click a badge to add that driver to the comparison, and click it again to remove them. Selected badges carry a ring. A retired driver, or one with no strategy in the snapshot, is greyed out and cannot be clicked. Until you click a badge, the widget compares the predicted top two. The choice belongs to this widget alone: the Timing Tower selection does not change it, and it resets when you change session. **Zooming.** Right-drag across the plot to draw a box, and on release the chart zooms both axes to it. The drag carries on if the pointer leaves the plot or the chart, and the zoom lands at the last point inside the plot when you let go. A right-click without moving opens a small menu with **Reset zoom**, which returns the chart to its opening view. Left-drag is unchanged and there are no zoom sliders. Below the chart, a footer reports: * how many unreachable points were dropped from the plotted curves, when any were; * that the rank-1 strategy is unreachable and so has no optimum marker, when that is the case; * an as-of stamp, when the snapshot behind the chart is stale relative to the playhead. *** ## Data Behind It Curves are built from the latest `strategy_snapshot`'s `sensitivity` rows, at or before the playhead lap (the as-of rule), filtered to the drivers picked on the badge strip. `model_id` `strategy_snapshot`, `model_version` the composite `engine_version+WRAPPER_VERSION`, over `vpw:v1:{session_key}:insight:strategy_snapshot`, hydrated via REST on session enter and rehydrate, then live. For each driver's rank-1 strategy and each stop index, `sensitivity` gives the predicted total time if that one stop moved to a given lap, with the other stops held at their optimum. A point at or above the engine's `1e300` unreachable threshold ("this stop lap cannot complete the strategy") is dropped from the curve and counted rather than plotted at a nonsensical value. The optimum marker is the driver's own rank-1 `predicted_time`; when that rank-1 strategy is itself unreachable, no marker is drawn and the footer says so instead. *** ## States > **No snapshot at all** > > Shows the shared empty-state text: the exact-lap status gate and reasons where known, or a generic "No strategy data at lap N." > **Snapshot present, no drivers selected** > > Shows "Select drivers above to compare stop sensitivity."; no drivers means no curves to draw. This happens once you have clicked every selected badge off. > **Drivers selected, no reachable points** > > If the reason is specifically that the rank-1 strategy is unreachable, shows "Rank-1 strategy unreachable; no optimum marker." Otherwise shows "No reachable sensitivity points." > **Stale snapshot** > > The chart still renders, with an as-of stamp in the footer: "Data as of lap N (playhead lap M)". > **Race-only** > > Outside a race, renders "Session is not a race"; the app never computes a strategy bundle for a non-race session. *** ## FAQ > **Why is there no optimum marker on a driver's curve?** > > That driver's rank-1 strategy is currently unreachable from the race state (the engine's `DBL_MAX` sentinel). The footer states this explicitly rather than omitting the marker silently. > **Why does a curve have fewer points than expected?** > > Some candidate stop laps cannot complete that strategy from the current race state and are dropped, not plotted at an invented value. The footer reports how many were dropped. *** ## Next Steps > **[Race Strategy Tool](/software/internal/watchparty/strategy/index)** > > The shared engine, derivation layer, and sentinel rule behind all six strategy widgets. > **[Strategy Table](/software/internal/watchparty/strategy/table)** > > The ranked strategies these curves are drawn from. > **[Predicted Pit Window and SC](/software/internal/watchparty/strategy/predicted-pit-window)** > > Each driver's next predicted stop, plus safety-car guidance. > **[Strategy Status](/software/internal/watchparty/strategy/status)** > > The gate, reasons, and whole-race health strip that explains a blank or stale chart. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.