Galvanic Polars: The Polar Diagram Your Boat Draws of Itself

Every new boat comes with a polar diagram. It is a beautiful curve, drawn in the back of the manual, that tells you what the manufacturer believes your boat should do in a given wind — eight knots upwind in twelve of true, six and a half on a beam reach in ten, that sort of thing. It is, for the day on which it was drawn, an honest document.

The day on which it was drawn was not your day.

First, What a Polar Diagram Actually Is

For the reader who has never lingered on one, a polar diagram is a deceptively simple object. On a sheet of polar paper, the angle from the centre represents the true wind angle (TWA) — zero degrees being straight upwind, one hundred and eighty being dead downwind. The radial distance from the centre represents the boat speed the hull is expected to achieve at that angle. One curve, drawn for one fixed true wind speed (TWS), traces out the locus of “how fast this boat goes, at this wind angle, in this wind.” A full polar is a family of such curves — one for six knots of breeze, one for eight, one for ten, and so on — overlaid on the same plot (Marchaj, Aero-Hydrodynamics of Sailing, 2nd ed., Adlard Coles Nautical, 2000; Larsson, Eliasson & Orych, Principles of Yacht Design, Adlard Coles, 4th ed., 2014).

Three working uses follow from the same curve. The first is velocity made good, or VMG — the component of the boat’s velocity in the direction of the wind, computed as Vboat · cos(TWA − target). The maximum of that quantity, upwind and downwind, gives the optimal angle to sail to make distance against (or with) the wind, a question every helm asks on every beat. The second is passage planning — given a forecast wind field over a route, the polar lets a router predict times of arrival and the optimal sequence of headings. The third is trim and tuning feedback — a deviation between observed speed and polar speed at the same TWA / TWS is, all other things being equal, a signal that something on deck is not quite right.

Where do polars come from? Historically, from Velocity Prediction Programs (VPPs) — numerical models, originally pioneered by Kerwin and others at MIT in the late 1970s, that solve a steady-state equilibrium between hull resistance, sail force, and heeling moment for every (TWA, TWS) pair. Modern handicap polars — most prominently those used by the Offshore Racing Congress’s ORC VPP — fit hull resistance data to the Delft Systematic Yacht Hull Series, a decades-long programme of towing-tank measurements run at Delft University of Technology (Keuning et al.). For a production cruiser, the published polar in the back of the owner’s manual is, in nearly every case, a sister-hull VPP output, computed for ideal conditions, with brand-new sails, an immaculate bottom, and no human factor anywhere in the equation.

The Polite Fiction in the Manual

The polar in the back of the manual was drawn for a sister hull, at the factory, on flat water, with brand-new sails, by a delivery skipper trying to make a deadline. It assumed a clean bottom, an empty galley locker, no reef in the main, no metre of swell, no current. It assumed the helm was held by somebody who has spent the past year coaxing similar boats through their press shots.

By the end of your first season, that polar is a polite fiction. The bottom has barnacles, and the published literature on biofouling — most prominently the work of Schultz at the US Naval Academy — shows that even light slime films measurably increase viscous hull resistance, with progressive calcareous fouling pushing the added drag sharply higher (Schultz, Biofouling, 2007). The genoa has stretched; sailmakers will tell you that the draft of a dacron headsail migrates aft and deepens within the first hundred hours of use, shifting the sail’s lift and drag coefficients away from the values the VPP assumed when it drew your curve. The mainsail no longer holds the same draft it did the day it was bagged. You have learned, on long passages, that the boat actually likes to be held a little lower than the brochure suggests upwind in a chop. You have a permanent half-litre of diesel in a place it should not be. None of this appears on the curve in the manual.

And on the day you actually want the polar — “is it worth bearing off ten degrees to Mahon to put the wind on the beam, or do I hold the rhumb line and pinch?” — you cannot use it. It is printed. It is frozen at one moment, by people who were not you, on a boat that was not yours, in a sea that was not the one outside your cockpit. It is blind to the reef you have in. It is blind to the swell. It is blind to the fact that you are motorsailing because the wind died at lunch. And it is blinder still to the fact that the same boat sails differently depending on who is at the helm.

A Polar Is Not Just a Property of the Boat

This is the part of the conversation that the brochure forgets. A polar is not a property of the boat. It is a property of the boat plus the people on board. The same production cruiser, with a delivery crew chasing a return ferry in the morning, sails very differently from the same hull on a Sunday afternoon with a five-year-old napping in the V-berth and the owner unwilling to put the rail under for anyone. The hull is the same. The sails are the same. The boat is not.

A polar that is honest about you has to reflect the way you trim. The angles you are willing to hold. The heel you are willing to live with at three in the morning in lumpy weather. Frank Bethwaite, in High Performance Sailing (Adlard Coles Nautical, 1993), spends a long chapter on exactly this — the gap between what a boat is capable of in the hands of a top crew and what the same boat delivers in the hands of its owner — and concludes that the gap is rarely a hardware problem. It is a behaviour problem. And it is one the brochure has no way of measuring.

If you want to know when you will arrive at Mahon, you should not be using a polar computed for the America’s Cup team. You should be using a polar computed for you, your crew, your sails, and the boat in the shape it is in this season.

What if the Boat Wrote Its Own Polar?

The idea of replacing a VPP-computed polar with an empirically measured one is not new in itself. Racing programmes have been doing it with dedicated logging stacks for forty years — Kerwin’s first VPP at MIT in 1978 was already calibrated against measured boat data, and modern America’s Cup syndicates carry instrumentation budgets larger than the entire price of a cruising yacht. What has been missing is the same idea applied — continuously, unobtrusively, and without a separate instrumentation install — to the boat that the owner actually sails on weekends and on the Mahon passage in August.

The principle behind Galvanic Polars is simple. Every minute the boat is underway, it already has access — over NMEA 2000 — to the data that a polar diagram is, in the end, the statistical summary of.

• True wind speed and true wind angle, fused from the masthead anemometer, the GPS, and the heading sensor.
• Speed over ground (and speed through water, where the boat has a paddle-wheel transducer that survived the last haul-out).
• Heel, pitch and roll, from the motion sensor — which, in turn, lets the boat infer the sea state (a half-metre chop and a flat reservoir produce very different polars at the same wind, and pretending they do not is one of the silent lies of the manual curve).
• Engine state, from the engine gateway — so that the boat knows when it is sailing, when it is motoring, and when it is motorsailing. The motoring records do not get averaged into the sailing polar; they are filed separately and used for what they are actually good for, which is detecting motorsailing later on.
• The sail configuration in use, declared by the owner once, in the app, against the boat’s actual sail inventory — main with one reef, main with two, full genoa, reduced genoa, staysail, code zero, asymmetric. Each record is filed against the configuration that produced it.

Every ten seconds, while the boat is moving faster than two knots, a filtered record of all of that is written to disk. The fusion of these heterogeneous sensor streams — apparent wind, GPS-derived speed, heading, attitude — into a single smoothed estimate of true wind and boat state is performed with a Kalman filter (Kalman, Trans. ASME J. Basic Eng., 1960), a recursive state estimator originally formulated for ballistic missile guidance and now standard practice wherever noisy multi-sensor data needs to be reconciled in real time. Every record carries a position, so the polar knows whether it was learned beating up the Solent or running across the Bay of Biscay. Every record is binned not by a single number but by a percentile band — at each TWA / TWS cell, the boat keeps a histogram of observed speeds, so the polar can report a median, an upper-envelope (95th percentile), and a confidence based on the number of samples in the cell. It never lies. It just shows less when it knows less.

After a season of sailing — sometimes after a single passage — what emerges is a polar that is unmistakably yours. This hull. These sails. This antifouling. This sea state. This crew.

Sea State — the Variable Nobody Uses

Among all the variables that move a yacht polar around, sea state is the most powerful one, and the one the manual is most quietly silent about. Two passages on the same boat, in the same true wind, in the same sail configuration, at the same true wind angle, can show boat speeds that differ by twenty, thirty, occasionally fifty percent — entirely because one was sailed in a flat anchorage and the other in a metre of swell on the nose. There is no part of the brochure polar that knows this. There is no axis on the diagram for it. There is no row in the table for it.

The physics is well understood. A hull moving through waves experiences an added resistance in waves (often written R_aw), distinct from the calm-water resistance the VPP starts from. Its dominant term scales roughly with the square of the wave amplitude, and becomes a substantial — sometimes the dominant — component of total drag once significant wave height passes about a metre. The classical energy-method formulation of this contribution is Gerritsma & Beukelman’s (International Shipbuilding Progress, 1972), with Maruo’s earlier theoretical treatment giving the upper-bound estimates still used as sanity checks today. Modern computations rest on strip-theory formulations originally laid out in Salvesen, Tuck & Faltinsen (Trans. SNAME, 1970) and elaborated extensively by Faltinsen in Hydrodynamics of High-Speed Marine Vehicles (Cambridge University Press, 2005).

And then, on top of the added wave resistance, there is the special case the modern flat-bowed cruiser knows very well: slamming. When a fin-keeled hull with significant forefoot flare pitches into the trough of a head sea and re-enters the water, the impact between the bow and the wave is not a smooth re-acquaintance — it is, in hydrodynamic terms, a sudden water-entry event whose impulse decelerates the hull, transfers a high-frequency pulse of energy into the structure, and bleeds kinetic energy that would otherwise have been propelling the boat forward. The theory of water entry was first laid down by von Kármán (NACA Technical Note, 1929) and refined by Wagner (1932), and the modern engineering treatment of slamming on planing and semi-displacement craft is again Faltinsen’s. The practical consequence for the cruising sailor is unromantic: every slam costs you boat speed, and a few slams a minute on a long beat will quietly dismantle the polar curve you thought you were sailing to.

Here is the paradox at the heart of all this. Most modern boats already carry the sensor that can measure sea state. The accelerometers and rate gyros embedded in modern AHRS-equipped autopilot heads, in motion-compensated wind sensors, and in the NMEA 2000 attitude and heave PGNs (127257 for attitude, 127252 for heave) are continuously emitting the very signal — heave amplitude, pitch variance, roll variance — from which significant wave height and sea magnitude can be inferred in real time. Decades of marine- instrumentation evolution have put a motion sensor on most production cruisers built in the last fifteen years. And almost nobody uses it for anything beyond stabilising a compass heading or trimming the heel readout. The data is on the bus. The polar pretends it is not.

Galvanic Polars treats sea state as a first-class axis. Every ten-second record carries a sea-state magnitude derived from the IMU’s heave amplitude and from the variances of pitch and roll over a one-minute window. The records are binned by sea state alongside true wind speed and true wind angle. When the polar is read back, it is read back at the sea state you are actually in — the flat-water curve and the metre-of-chop curve are separate objects, and the boat shows you the one that matches the cockpit you are sitting in. If you have only ever sailed your boat in flat water, the rough-water polar will say so honestly. It will not paper over the gap with the calm-water curve.

Stored as a Ratio, So It Scales Honestly

One detail that matters, and that is rarely talked about. The Galvanic polar does not just memorise “in 12 knots of true wind at 60 degrees apparent, this boat does 7.1 knots.” It memorises the ratio between boat speed and wind speed at every angle — and stores it in a way that lets the boat reason about wind it has not specifically seen yet.

Which means: when tomorrow brings 14 knots instead of 12, the boat does not shrug and say “no data for that exact condition.” It scales the curve from the wind it has learned, presents the most relevant slice, and tells you plainly how confident it is. The principle is the same one that underlies the use of non-dimensional groups in naval architecture — the Froude number, the Reynolds number, the speed-to-length ratio — that make towing-tank data scalable to full-size hulls in the first place. A polar normalised by the true wind speed inherits that scalability property and lets a season’s worth of measurement at one set of windspeeds remain useful at another. (This particular method — what we call the ratiometric speed-normalisation polar — is one of the embodiments in our pending patent portfolio.)

Today First. “What If” on a Slide.

The other thing the polar in the manual gets wrong is the display itself. It shows you every wind, all the time, on the same chart — which means that the wind you actually have right now is one of a dozen competing curves on the page. Useful at a workshop. Useless at the wheel.

Galvanic Polars, in the app, shows the slice that matches today’s wind first. The current heading is highlighted on the curve. The optimal-VMG angle toward your selected destination is drawn on top of it. The display defaults to the wind range you have actually been sailing in for the past hour, because that is the question you actually have at the helm.

And if you want to look at something else — “what about twelve knots? What about beating in eighteen?” — you slide. The slider opens the what-if view; the boat replots for the selected condition. When you stop touching the screen, it quietly reverts to now, because you have other things to watch. (The wind-angle compass with the integrated speed heatmap — what you see on the screen when you open the Polar tab — is another of the pending patent embodiments.)

What Stops Being a Guess

Once the boat has been writing its own polar for a few weeks, a small list of questions stop being guesses and start being readings.

• “Is it worth bearing off ten degrees to put the wind on the beam, or should I hold the rhumb line and pinch?” — the answer is on the screen, in your data, against your actual rate of progress at both angles.
• “Am I getting the most out of this boat right now?” — the deviation from your own historical median is shown as a percentage. Not against the brochure. Against yourself.
• “Am I motorsailing?” — the engine gateway and the polar know between them. The boat can say so, calmly, without you having to admit it.
• “How much earlier would I get there if I trimmed properly?” — the answer is the distance between where today’s point sits on the heatmap and where the upper envelope of your own historical data sits.
• “When will I arrive at Mahon?” — answered against your polar, in your sail configuration, in this sea state, not against a curve the factory drew before you ever stepped aboard.

After Some Training — An Honest ETA From A to B

After a season — sometimes after a passage or two — of letting the boat write its own polar (the user manual covers the training period in detail, including how to read the confidence indicators while the polar is still maturing), what you have on the screen is the tool that answers the single question every sailor asks at the start of every passage: when will we arrive at B?

Not the answer for a racing crew in this boat with new sails on a flat sea. Not the answer the brochure gave you. Not the answer the chartplotter computes from a generic factory polar it imported once and never updated. When will you arrive — with this hull, these sails as they are this season, the way you trim, the heel you are willing to hold, in the sea state outside the cockpit right now, given the wind the forecast is offering you over the next twelve to forty-eight hours.

The answer is built by composing three honest inputs: (a) the empirical polar of your boat, learned over your own sailing time; (b) the sea-state magnitude the IMU is reporting right now, which the polar is read against; (c) the wind forecast over the route from A to B. The result is a passage time that is yours — not a baseline copied off a sister hull.

And — perhaps the point that matters most — the same answer is useful to everyone on board, whatever the boat is being used for that weekend.

• The racer uses it as the honest baseline to measure tactical decisions against. “If I bear off ten to put the wind on the beam, do I gain or lose on the rhumb line to B?” — answered against your own data, not a borrowed curve.
• The cruising sailor uses it as the honest answer to “will we make Mahon before dark, or do we need to start the engine at four?” — answered, today, in the wind and the sea state you actually have.
• The slow sailor uses it for what slow sailors need most — knowing, an hour in, whether the passage as planned still gets them into harbour at a reasonable hour, or whether to revise the destination while there is still light to do it.

The polar is the same polar in all three cases. It is the polar of this boat, with this crew, in this sea. Galvanic Polars makes it available — and lets it answer, at last, the question the brochure curve never could.

The Boat in the Brochure Has Nothing to Teach the Boat That Just Crossed the Med Twice

The polar you actually use is the one your boat — and your crew — have drawn between them, over the passages you have actually sailed, in the wind you have actually had, with the sails you have actually deployed.

It is a small thing, and it is not. A polar honest about you is the difference between an estimate of arrival that is a round-number guess and one that is a measurement; between trim that feels right and trim that is right; between asking the manual what the boat ought to do, and asking the boat what it has been doing all along.

The brochure boat and the just-crossed-the-Med boat are different boats. Galvanic Polars is for the second one.

References

1. Marchaj, C.A. Aero-Hydrodynamics of Sailing. 2nd ed. Adlard Coles Nautical, 2000. (The standard reference on the physics of a sailing yacht in the wind and the water; the source for the geometry and meaning of the polar diagram used here.)
2. Larsson, L., Eliasson, R. & Orych, M. Principles of Yacht Design. 4th ed. Adlard Coles, 2014. (Naval architecture textbook; covers Velocity Prediction Programs, hull-resistance modelling, and the limits of steady-state polars.)
3. Bethwaite, F. High Performance Sailing. Adlard Coles Nautical, 1993 (revised editions thereafter). (Empirical sailing science; extensive treatment of the gap between what a hull is capable of and what its crew extracts from it.)
4. Kerwin, J.E. A Velocity Prediction Program for Ocean Racing Yachts. MIT Department of Ocean Engineering, 1978. (Among the earliest published VPPs; the methodological ancestor of the polars in the back of every modern production-boat manual.)
5. Keuning, J.A. et al. The Delft Systematic Yacht Hull Series. Delft University of Technology — a multi- decade programme of towing-tank measurements used as the hull-resistance basis for most modern VPPs, including the ORC VPP.
6. Offshore Racing Congress. ORC VPP Documentation. Annually updated, publicly available at orc.org. (The Velocity Prediction Program behind ORC handicapping; instructive on how published polars are computed and what they assume.)
7. Schultz, M.P. “Effects of coating roughness and biofouling on ship resistance and powering.” Biofouling, 2007. (The widely-cited US Naval Academy work on the measurable cost of even light hull fouling.)
8. Gerritsma, J. & Beukelman, W. “Analysis of the resistance increase in waves of a fast cargo ship.” International Shipbuilding Progress, 1972. (The classical energy-method formulation of added wave resistance — the variable that dominates real-world polar scatter once the sea-state climbs.)
9. Salvesen, N., Tuck, E.O. & Faltinsen, O. “Ship motions and sea loads.” Transactions of the Society of Naval Architects and Marine Engineers (SNAME), 1970. (Strip-theory framework still underlying most modern ship-motion and added-resistance computations.)
10. Faltinsen, O.M. Hydrodynamics of High-Speed Marine Vehicles. Cambridge University Press, 2005. (The modern engineering treatment of slamming pressures, water-entry impacts, and added resistance in waves.)
11. von Kármán, T. The Impact on Seaplane Floats during Landing. NACA Technical Note 321, 1929. (The foundational analysis of hydrodynamic water-entry — the physics behind every bow slam since.)
12. Wagner, H. “Über Stoß- und Gleitvorgänge an der Oberfläche von Flüssigkeiten.” Zeitschrift für Angewandte Mathematik und Mechanik, 1932. (The wedge-entry refinement of von Kármán’s impact theory; still cited in modern slamming-pressure calculations.)
13. Kalman, R.E. “A New Approach to Linear Filtering and Prediction Problems.” Transactions of the ASME — Journal of Basic Engineering, 1960. (The recursive state estimator used to fuse the wind, speed and attitude channels before any polar record is written to disk.)