Tacking Cost, Measured

This is a measurement piece, so the method matters as much as the answer. Here is what the published guidance claims, what the tracks actually show, what the data genuinely cannot tell you, and what the whole thing means the next time you are deciding whether to put in one more tack.

What the guidance says

The industry numbers are remarkably consistent, and remarkably vague. Sailing World puts a good tack at "somewhere between a half- to a full-boat length, at a minimum," and frames the race in terms of how often you do it: six, seven, eight times. North Sails says the difference between a good tack and a poor one "can be measured in boat lengths," and that in a race where you tack ten times, good tacks can be the margin of victory.

Both are talking about a well-drilled crew doing it properly. Neither is a measurement of what happens on a Thursday night at a club, which is the thing most of us actually want to know.

The races

Two races on Utah Lake on the second night of the Fall Breeze Series, the same evening covered in our night-two recap. Four Catalinas: two Catalina 25s, a 27 and a 28. One lap of a short triangle, with a first beat of just 370 metres from the middle of the start line to mark 1. Every boat carried a phone recording its position, and the committee scored the night on the light-air time-on-time constants, which tells you what the evening felt like before you look at a single number.

It was slow. Across the fleet, steady close-hauled boat speed on that first beat came out at 2.75 knots. This is a drifter, and that matters for everything that follows.

What GPS can and cannot see

This is the part most analyses skip, and skipping it is how you get confident numbers that are wrong.

The tracks record one fix every 15 seconds. A keelboat tack takes somewhere between ten and twenty seconds from the moment the helm goes down to the moment the boat is back up to speed on the new tack. So there is at most one position fix inside the manoeuvre, and often the tack straddles a single 15-second interval with no fix inside it at all.

That rules out the thing you would most like to do, which is plot the speed curve through the turn. You cannot see the deceleration, the head-to-wind moment, or the acceleration out. Anyone showing you that curve from 15-second data is drawing it, not measuring it.

What you can do is measure the interval that contains the tack. If a boat's steady close-hauled speed is 2.75 knots and the 15 seconds containing a tack averages 1.01 knots, the difference multiplied by the interval is distance that tack cost, and it is a real measurement. It is also a floor: the loss is smeared across the whole interval, and any part of the manoeuvre that spills into the neighbouring interval is invisible.

The measurement

Across both races there were four tacks that fell cleanly inside a single fix interval, on three different boats. Comparing each one against that boat's own median steady close-hauled speed on the same leg:

So a tack on these boats, on this night, cost about nine seconds and one and a half boat lengths. That is roughly double the published figure for a well-executed tack, and the reason is sitting in the speed column: at 2.75 knots there is almost no momentum to carry a cruiser-racer through the turn. Half a boat length is what you achieve with a drilled crew and enough breeze to coast. In a drifter you stop, and then you have to start again.

The consistency is worth noticing too. Four tacks, three boats, three different hulls, and every one landed between 7.7 and 9.6 seconds. For a sample this small that is about as tidy as field data gets.

And now the part that actually decides races

Here is the beat those tacks happened on. The rhumb line from the middle of the start line to mark 1 was 370 metres. This is how far each boat actually sailed to get there, and how long it took:

Summer Breeze finished that leg 169 seconds behind Otter Spotter. It made one more tack than Otter Spotter did. That extra tack cost it, by the measurement above, about nine seconds.

Nine seconds out of a hundred and sixty-nine. Five per cent.

The other 160 seconds are distance. Summer Breeze sailed 301 metres further than Otter Spotter did to reach the same mark, and at the fleet's steady 2.75 knots, 301 metres takes about 213 seconds to cover. The extra distance does not partly explain the deficit. It explains essentially all of it.

True Luff makes the point even harder. It tacked the same number of times as the leg winner, so its manoeuvre cost was identical. It still lost 313 seconds, because it sailed 447 metres further.

What to do with this

The instinct this data should kill is the one where you sit on a tack you know is wrong because tacking is expensive. It is expensive. It is nine seconds. You will give away nine seconds twenty times over by spending four minutes going the wrong way.

• A tack that puts you on the correct side is always worth it. On this beat, one extra tack cost five per cent of what one wrong decision about which side to sail cost.
• Count your tacks, but weigh your distance. If you are sailing much more than about 1.4 times the rhumb line on a beat, the problem is not your boat handling. That ratio is roughly what a 90 degree tacking angle costs you, and anything well beyond it is course made good going to waste.
• Light air makes the tack dear and the mistake dearer. Both costs scale the same way: everything takes longer at 2.75 knots. The ratio between them does not change.
• Practise tacks anyway. Nine seconds is real, and over the six to eight tacks of a normal race it is a minute. It just is not the first place to look when you lost a leg by three.

Measure your own

Nothing above needed anything other than the race's own data. If you want to run the same numbers on your fleet, you can export a race as a set of CSVs with every boat's track, the marks, the finish times and the results, and hand it straight to an AI assistant with our free Race Analyst skill, which knows the file format and the cleaning rules.

One warning from doing this exercise: raw GPS is messier than you expect. One boat in this fleet dropped its track for nearly six minutes and came back with a fix that looked like a manoeuvre but was actually a gap. Its two apparent tacks are excluded from every number in this article for exactly that reason. If you measure your own races, look at the time gaps between fixes before you believe anything the positions tell you.

How much to trust this

An honest summary of the limits, because a number with no error bars is just a stronger opinion.

• Four tacks. Three boats, one evening, one venue. This is a measurement, not a study. It tells you what happened on that beat and gives you an order of magnitude, nothing more.
• One wind strength. 2.75 knots of boat speed is the bottom of the range. In 12 knots the seconds would be different, and probably smaller.
• Cruiser-racers, not one-designs. Four Catalinas between 25 and 28 feet with club crews. A drilled one-design crew tacks better than this, which is exactly what the published half-to-one-boat-length figure is describing.
• The method gives a floor. Smearing the loss over a 15-second interval understates any part of the manoeuvre that fell outside it, so the true cost is this or worse.

What survives all of that is the comparison, and the comparison is not close. Whatever a tack costs on your boat, it is a small number of seconds. Whatever sailing the wrong side of a beat costs you is minutes. Spend your practice time on the first and your thinking time on the second.

Sources

1. Race tracks, marks, finish times and results: the Committee Boat HQ race records for Fall Breeze Series #2 and Race 2a, Utah Lake, September 10, 2026. All figures in this article were computed from those exports.
2. "Three Essential Tacking Tips", Sailing World (a good tack costs between half and a full boat length at a minimum; six to eight tacks in a race).
3. "How to Improve Your Tacks", North Sails (the difference between a good and a poor tack measured in boat lengths; the phases of the manoeuvre; faster turns needed in waves).
4. Won at the Gun: Fall Breeze Series, Night Two, Committee Boat HQ (the full recap of the two races the tracks come from).
5. Reading Your Own Race Replay: 5 Mistakes You Can Only See From Above, Committee Boat HQ (what a 15-second track can and cannot show you).