What a calibration is for
A polar says what the boat should be doing at a given true wind angle in a given breeze, and that comparison does not survive an instrument a few degrees out. A masthead unit bolted down three degrees off makes every pointing angle, every wind direction and every percentage of target wrong by the same three degrees, until somebody touches it. Calibration here has one purpose: a wind number the polar can be compared with.
The order is compass, boat speed, then wind
Nothing on a boat measures true wind. TWA, TWD and TWS are computed, from the masthead unit and also from heading and speed through the water. So the order is fixed: compass, then speed log, then wind. A log reading two per cent slow puts an error into every wind number the boat produces, and a wind correction measured through it carries that error into the entry you save.
- Heading, against course over ground. Motor in flat water with no current and read the compass against COG on several headings rather than one: a compass is usually wrong by a different amount on different bearings.
- Boat speed, against speed over ground. Across the speed range and at the heel angles the boat races at: a paddlewheel that reads true upright and slow at twenty degrees of heel is ordinary.
A correction is a lens, not an edit
An entry is linear — value × scale + offset, scale first — and reaches forward from an instant you give it, stated in UTC. On an angular channel it is a rotation and nothing else, so the scale field is removed rather than greyed out: a scaled angle is not an angle. The calibration lens is the full argument.
What guided calibration measures
Analyze reads a window of your own racing. It stores nothing, and an Apply writes the same ordinary entry you would have typed by hand.
It rests on one sentence: a wind shift is attached to the clock, and a constant angle error is attached to the tack. The analysis cuts the day into steady close-hauled segments, takes the median wind angle of each and then the median of those per board, and half the difference is the offset. The same number goes to true wind angle and direction — and deliberately not to apparent wind, where the wind triangle would make it the wrong correction.
Three things it declines to do:
- It does not measure below its own evidence. Four tacks and three steady segments per tack are the floor, and below the noise floor nothing is suggested.
- It does not separate the vane from the day. The constant is where the masthead unit sits plus that day’s wind shear, and one window of sailing cannot tell the two apart. So analyse a day at a time, and date each entry to it.
- It does not read boat speed as an instrument error. The ratio of GPS speed to boat speed is reported and never recommended: current flows straight into it.
A crew can imitate a constant even so: a board sailed lower and faster carries a wider wind angle on every beat. So each finding carries what both boards held — accepted time, median wind angle, median boat speed — with a caution past 0.20 knots between the two speeds.
The cross-check reads wind direction either side of every tack: true wind direction belongs to the water, so a step there was produced by the boat. A step inside its own spread confirms the finding. A real step warns that part of the error changes sign with the tack, which no linear entry corrects — and it has no Apply button.
On synthetic data, an injected 3.0° vane error is recovered as −3.04°, and a re-run through the applied entry measures a residual of −0.02° and suggests nothing.
Why a vane’s error can vary with the wind
Not everything a vane gets wrong is a constant. The sail plan bends the air before it reaches the masthead, so the vane reads a wider angle than the boat is really sailing — on both tacks, which is why it lands as a step across the tack rather than as a difference between the boards. That is upwash: largest in light air, smaller as the breeze builds, and able to push the other way under backstay load, as mast bend twists the masthead unit to windward. A mounting offset cannot produce a step across the tack, and a table that widens both tacks equally cannot produce one, so each is invisible to the other’s measurement.
How the table is derived
The upwash profile reports that step resolved by wind speed — one row per fixed 2-knot band, tacks and gybes kept apart. A row becomes a cell in two operations.
Halve the measured step, and turn it around. Halve it because the step is the full difference between the boards, twice the error on either; turn it around because a positive step means the vane reads wide. +3.6° in the 6–8 knot band becomes −1.8° in that band’s upwind cell.
A band’s number is read as a measurement at its midpoint, and between two populated midpoints the correction is interpolated linearly in wind speed. Outside that range it is held flat: nothing extrapolates, so a table measured to 16 knots serves its 16-knot number in 30. Under three manoeuvres a band is hollow — its count and no number — and the prefill leaves that cell empty. Upwind applies at or inside 65°, downwind at or beyond 120°, the reach between them a cross-fade.
The limits are published with it: one profile per boat rather than per sail configuration, no correction of apparent wind angle or of wind speed, a reach interpolated and never measured, sea state, heel and crew unmodelled, and a cell beyond ±20° refused.
Re-analysed through a table built this way, the same synthetic day took the sum of the absolute steps across the bands from 31.28° to 0.86°, and left the tack-symmetry suggestion unmoved.
Where each of these lives
Open Boats and take Calibration from the boat’s row, beside Polar and Targets. Admins and owners write, every role reads, and the page is never billing-locked. It runs in the order of the work: Analyze, with the findings, the cross-check and the upwash profile under it; the Wind angle correction table editor; then Add a correction over the timeline of what is in force.
Use as a correction table fills that editor from a measured profile, halved and negated, and stops; the editor never saves as you type, and Save as a new correction appends exactly one entry. In the debrief a cal badge rides the data wherever a per-channel number is drawn — the Explorer’s strip heads and the tuning guide’s tiles.
A correction table is the boat’s own measured signature rather than a setting we recommend, and it is not portable: how much of it a boat carries depends on where its vane sits.
Read the method
The documentation this post draws on.
- The calibration lens — Why a correction is applied when data is read rather than written into it, and the commutation that lets it apply to a whole season instantly.
- Guided calibration — Measuring a wind instrument’s constant angle error from your own sailing — the physics, the algebra, what that constant contains (the mounting offset and the day’s wind shear), and the cross-checks that keep a suggestion accurate.
- The upwash correction table — Correcting a wind-angle error that changes with the breeze: fixed 2-knot bands, half the measured step and never the step, and interpolation that clamps rather than invents.
- Calibration in practice — Entering a correction, measuring one from your own sailing, reading what Analyze found, applying it — and undoing any of it exactly.