Tack Choreography: Turning Without Losing Speed

A fast tack is a rehearsed sequence, not a single action — the driver turns the boat at a controlled rate while the trimmers, grinders, pit and crew weight all move in time, so the boat comes out of the manoeuvre and rebuilds to target speed with the least distance lost. Because you tack many times up every beat in upwind mode, the difference between a clean tack and a scrappy one compounds into real places. This is technique work, and it rewards drilling more than talent.

The turn is nearly free; the re-acceleration is the bill

Every tack divides into two phases, and understanding which one actually costs you reframes the whole manoeuvre. Through the turn, the hull swings its bow across the wind and trades forward momentum for windward displacement — you frequently come out of the turn slightly ahead in VMG of where you would have been on the old course. The loss almost entirely accrues in the re-acceleration: once the sails have crossed and the boat has slowed, it must be rebuilt from reduced speed back up to target, and that climb is expensive for two physical reasons.

First, hydrodynamic resistance scales with the square of boatspeed. Total drag on a displacement keelboat is the sum of viscous drag (skin friction and form drag), wave-making drag, and induced drag from the sideforce the keel must generate — and all of the speed-dependent terms rise roughly as v². That non-linearity is why the last knot back to target costs vastly more than the first: near the bottom of the slow-down the water barely resists you, but as you approach full speed the resistance you are pushing against is at its highest. The re-acceleration curve is therefore steep at the start and flattens badly at the top, which is exactly the region a mistrimmed sail or a high heading can stall you out of.

Second, accelerating the boat means accelerating its added mass — the volume of water entrained around the hull, keel and bulb that must be dragged along with the boat any time its velocity changes. Added mass is an inertial penalty that only appears under acceleration, and on a 3.25-tonne boat carrying a 1,100 kg bulb on a deep fin it is a genuine fraction of the effective mass you are trying to get moving. It is one reason a heavy keelboat feels "sticky" out of a tack in a way a dinghy never does, and it is why the manoeuvre punishes any wasted moment between the turn completing and full drive being restored.

The practical lesson: a textbook turn wrapped around a slow, uncoordinated exit is still a slow tack. The whole point of choreography is to compress the re-acceleration phase — sails setting cleanly, the boat flat, and the driver free to steer for speed the instant the turn completes.

Entry, turn rate and the apparent-wind case for a low exit

The turn follows a rate profile, not a constant rate. The standard shape is bear away a couple of degrees first to load the foils and bank a little reserve speed, then feed the helm in progressively: relatively slow up to head-to-wind, quickest through the eye where the sails are luffed and generating no drive anyway, then easing the rate off as the sails fill on the new side so the hull is not thrown into a skid. On a tiller-steered boat — the Melges 40 is steered by tiller — this is felt directly as helm load, which is precisely why tiller feedback suits fine speed steering: the driver reads the building sideforce through the stock and modulates the rate against it rather than against a compass.

Rudder angle is not free. A spade rudder held at a large deflection is a high-drag, high-lift device operating near its own stall; too much helm scrubs speed and, past roughly the mid-teens of degrees of angle of attack, the blade stalls and stops turning the boat efficiently. That is the mechanism behind the two classic driver errors: turning too slowly bleeds speed while the boat sits head-to-wind making no VMG, and turning too fast over-deflects the rudder, stalls the blade, and slews the hull sideways so the keel and rudder both lose their grip on the water. The fastest usable turn belongs to medium air and chop, where you can afford to be aggressive; flat water, light air, and heavy-air over-power all call for a slower, gentler rate.

Exit angle is set by the breeze, and the reason is apparent wind, not superstition. At full speed the boat's own velocity drags the apparent wind forward and strengthens it, letting the sails work at a small angle of attack while pointing high. The instant you finish a tack you have lost that boatspeed, so the apparent wind has swung aft and gone soft — and a close-hauled heading now presents the sails at an angle of attack too fine to drive. Exit low and you realign the sails to the temporarily-aft apparent wind, keeping them attached and pulling; as speed builds, the apparent wind marches forward again and you squeeze up to full height to follow it. A common working range is roughly five degrees low of close-hauled in light air, about three in moderate, and near on-course in heavy air where you already carry enough speed to point. Pointing high before speed returns is the self-inflicted classic: the foils never develop their working lift, the boat crabs sideways at a large leeway angle, and the acceleration drags on for extra lengths. Calibrate these numbers against your own polars and target boatspeeds — the exact angles belong to the boat, not to a rule of thumb.

The roles, timed as one

Coordination is the whole game. Each person has a defined job and a defined moment, and the sequence runs to a single countdown from the driver — typically a warning ("ready") and an execute ("tacking" or "helm's a-lee").

• Driver — owns the turn rate and the timing. Turns fast enough to be efficient but smooth enough never to stall the rudder, and the instant the turn completes stops steering the compass and starts steering for speed and heel, feeding the bow up only as fast as the apparent wind allows.
• Release trimmer — blows the old jib sheet at the instant the sail just begins to back: too early and the jib flogs uselessly through the foretriangle, too late and it backs hard against the shrouds and acts as an air brake right when the boat can least afford it. This release timing is one of the largest hidden variables in tack quality, because a backed jib not only brakes the boat but pushes the bow off and forces the driver to fight it with more rudder.
• New trimmer and grinders — take up the incoming sheet from zero. The grinders wind at full speed before the load arrives, so the sail is most of the way in as it fills and the winch is not asked to drag a fully loaded sail the last metre from cold. The final few centimetres to the target trim number are found on the fine-tune once speed is up and the leech load is readable.
• Pit — manages the runner exchange (below) plus any halyard, traveller or lead task, and clears lines so nothing snags mid-manoeuvre. On this boat the pit is also, in effect, running a mid-tack rig tune.
• Crew weight — crosses late and hikes immediately (see crew weight and hiking) to stand the boat up so it accelerates flat. With a maximum crew weight around 750 kg across nine to ten hands, that mass is a major share of righting moment and its timing across the boat materially affects how quickly the boat settles onto its lines. Housekeeping comes after the boat is settled, never during.

In light and moderate air the weight movement is not just relocation but a roll tack: heel to leeward first to induce weather helm and start the bow turning up, hold the crew in as the boat heads into the eye to bank angular momentum, then throw weight across to the new leeward side through irons to help drive the sails over and spin the boat around the crew, before coming up to hike flat on exit. The physics is conservation of angular momentum — storing rotational energy on entry and releasing it through the eye — and it is why a well-timed light-air tack can cost almost nothing while a flat-footed one stalls dead in the water.

Runners and the canard: two rig-tune events inside the manoeuvre

The Melges 40 carries no permanent backstay; the square-topped mainsail head clears where a standing backstay would run, so mast and headstay support come from topmast running backstays. That makes the runners far more than mast props — they set headstay tension, and therefore forestay sag and jib entry. Grinding runner load on tightens the headstay (flattening the jib entry and firming its leech), bends the mast to flatten the mainsail and open its upper leech, and generally de-powers the top of the rig; easing it does the reverse. So the runner exchange during a tack is a rig-tune event, not an afterthought once the boat is up to speed.

Practically, the pit or runner hand must ease the old (now leeward) runner and load the new (windward) one to the correct tension as part of the count. Get it wrong and the headstay is soft for the first, most speed-critical part of the new tack: the forestay sags to leeward, the jib entry goes round and powered-up, and the trimmer is fighting a sail shape the boat never asked for at exactly the moment they are trying to accelerate. On a boat this size the runner loads are high, so the exchange earns its own dedicated hand and its own place in the sequence. The specific loads, purchase ratios and any reference marks are boat-specific — verify them against the class rules and the boat's own rig documentation rather than trusting a generic figure.

There is a second, less obvious rig-and-foil event. Because the deep fin cants up to 45 degrees to windward for righting moment, it no longer sits on the centreline and generates less pure lateral resistance; the boat is understood to carry a forward canard on the centreline to make up the lost sideforce and hold leeway in check. Where a canard is fitted and adjustable, its state through the tack matters — the boat swaps which side is loaded, and the foils must be working from the moment drive is restored, or the low-speed exit crabs badly. The canting-keel actuation itself is a single Cariboni hydraulic ram on a 24 V, roughly 4,500 W pack, and canting the keel is a deliberate, powered move rather than a snap action, so on a straightforward tack the keel is typically settled either side of centre rather than swung live through the turn. Treat the canard's presence, geometry and control — and the twin-rudder configuration this boat is generally reported to run — as boat-specific details to confirm against the class rules and the boat's own documentation; sources are not fully consistent, and getting the setup right depends on the actual foil and rudder fit.

What good and bad look like

A good tack is quiet. The turn is one continuous, rate-controlled motion; the jib crosses without backing and is sheeted most of the way home before it loads; the runner is on and the headstay firm; the boat is flat and settled within a length; and the driver is already working the bow up to speed while the crew hikes. Boatspeed is called through the whole exit so everyone knows the instant target is regained.

A bad tack announces itself, and each failure mode has a physical signature:

• Turning too slowly — the boat sits head-to-wind bleeding speed while making no VMG, and the re-acceleration starts from a deeper hole.
• Turning too fast — the rudder over-deflects and stalls, the hull skids sideways, and the keel and rudder lose grip; leeway spikes and the exit is a crab, not a drive.
• Releasing the old sheet late — the jib backs against the rig and acts as an air brake, pushing the bow off and forcing more corrective helm.
• Grinding a cold winch — the sail arrives on the new tack half-trimmed and powerless, so drive is missing at the exact moment the v² drag curve is steepest and speed is hardest to build.
• A botched runner exchange — the headstay stays soft, jib entry goes round and powered, and the trimmer chases a shape the boat did not ask for.
• Pointing too high on exit — the foils never load to their working lift, leeway runs large, and the acceleration drags on for extra lengths.

Any one of these turns a two-length loss into a five-length loss — and these are precisely the common speed killers that separate crews of equal boat speed.

Drilling it

Choreography is muscle memory, so it must be rehearsed under load, not talked through on the dock. Isolate the manoeuvre: run tacks at a fixed interval in flat water and call boatspeed before and after each one so the loss is a number, not a feeling. Two metrics matter — target speed minus exit speed (how deep the hole was) and time to regain target (how fast you climbed out) — and both should trend down as the crew tightens. Rotate roles so every hand understands the full sequence and, critically, the handoffs between them, because most botched tacks are handoff failures, not individual errors. Film from directly astern to audit turn rate, jib-release timing and the exact frame the weight crosses; the camera catches the lag the crew never feels in the moment.

Drill across the breeze range you expect on the day, because a light-air tack (more roll, weight used to help the turn and drive the sails over, a lower exit) and a heavy-air tack (flatter, crisper, a smaller, gentler turn to protect the rudder, an on-course exit) load the boat and crew in almost opposite ways. Log the numbers in a boat-speed debrief so the improvement is visible over a season.

Tack choreography is efficiency under repetition: turn on a controlled rate to protect the foils, release and grind on time so the sails are drawing as they fill, exchange the runners and set the canard as a rig-tune, move as one and rebuild speed on a low, apparent-wind-aware exit. It is the twin of the gybe, and a little saved on every tack — measured in the shape of that steep v² re-acceleration curve — adds up across a beat, and across a regatta.