Good Sailing Performance?

[Colvic Watson]

I think speed is a relative thing. In light airs it's a bit pointless trying to point a boat like ours, but off the wind in 7 knots of wind 2 knots through the water is good enough if we're out for the day. I wish we had more canvass for 15 knots of wind then we could guarantee no engine for a passage but for a day sail, who cares? I don't get the need for absolute speed on a passage. You either get there sooner and spend more time in the marina - why? Or go further and miss the interesting places to stop en route.

It depends what you use the boat for, we use ours for day sailing and month long adventures, in either case speed is largely irrelevant. But on passages we always wish we were a knot faster, we're under canvassed and a passage with a tight tide gate is when it's a bit annoying.

 

[doug748]

“Good Sailing Performance….” - What does this term really mean?"


I think I know what it means to some. There is no shortage of the Overtaking Anecdote they crop up all the time and often take the general form:

- "My boat once sailed past another boat" - We are invited to understand three things from this episode:


* My boat is a "faster" design than the other boat.

* My boat therefore has a "better sailing performance"

* This proof shows my boat is "better" than the other boat

Very creaky logic in my view.


For the sort of sailing you envisage I would go for a boat that will sail in light airs but will not need a lot of sail tending. A boat that will not overpress it's helm easily, in lively conditions, and will be light work shorthanded. "Speed" in itself is a red herring, provided she tacks cleanly, heels to the wind and carries a light helm with a good spread of sail. You also need to keep your expenditure within bounds. No point in putting your 30k in the bank only to take most of it out again.

I might forego the Contessa (they are fine but often rather pricey) and seek a fin keeled cruiser of the Trapper, Sabre, Sadler 25 type. If you look for the bargain rather than the specific design you might do very well for as little as 6k, which will please the missus and still give you a capable, fun boat.

 

[Neil]

“Good Sailing Performance….” - What does this term really mean?"

For the sort of sailing you envisage I would go for a boat that will sail in light airs but will not need a lot of sail tending. A boat that will not overpress it's helm easily, in lively conditions, and will be light work shorthanded. "Speed" in itself is a red herring, provided she tacks cleanly, heels to the wind and carries a light helm with a good spread of sail. You also need to keep your expenditure within bounds. No point in putting your 30k in the bank only to take most of it out again.

I might forego the Contessa (they are fine but often rather pricey) and seek a fin keeled cruiser of the Trapper, Sabre, Sadler 25 type. If you look for the bargain rather than the specific design you might do very well for as little as 6k, which will please the missus and still give you a capable, fun boat.

This was my logic - I knew Herself would have very little interest, so I pleased myself with a boat that had (by reputation) a good turn of speed, easy to handle single handed, forgiving to sail, and for it's length, a good sea boat. I bought a Sadler 25 - cheaper than the 26 and can be picked up quite cheaply, and being small, incurs lower mooring charges maintenance etc, yet is still big enough to go on longer passages with a mate or two (but no more!). No hot water, shower or other Herself-friendly conveniences required. So far, It's lived up to expectations. Of course , I paid a premium for one with a brand new engine and all the works, new Furlex, toilet etc, and then compounded the 'error' by buying new sails, spray hood, GPS plotter, AIS radio, PLB etc etc......next, new instruments, standing rigging...well you know the drill. I could have bought something bigger and newer for the the money I've shelled out, but would I have been happier? probably not, and would still have wanted new sails, engine, blah-di-blah

 

[Mrnotming]

Leinster Boats in Dublin has a Ruffian 8.5 on the books.These are as tough as old boots and had/have a decent interior.For a non-Racer they bring you home every time, in a reassuring fashion.P& O have great rates for exporting her over with the Guinness!
No connection, just saw the ad!(Her/it's name is "Archie")Lying National Yacht Club Dun Laoghaire when I last saw her.Pretty scoop added.

 

[Tranona]

I don't buy it I'm afraid. Unless I've misunderstood they did a simple linear regression. This is obviously inappropriate as the speed is zero bounded, and also displacement hull speed is known to be proportional to the square root of the waterline length - which is pretty good hint that you're not going to get away with pretending the relationship is linear. Having done a linear regression on non-linear data they end up with an intercept of 2.62, which is completely meaningless, but they make a half-assed attempt to explain it. In all, the passage you quote looks to me like complete statistical twaddle, and I'm surprised that it is published.

The maximum displacement hull speed of any hull in knots is roughly equal to the square root of the waterline length in feet times1.34 or thereabouts. (With the exceptions of very narrow or wave-piercing hulls, and planing hulls.) This means that a lwl of 20 feet can just be pushed at a MAXIMUM of six knots. (OK, if you apply an absurd amount of energy you might be able to make a non-planing hull plane, but something would probably break.) The hydrodynamics of why this is the case are well understood. I plan on 4 knots for my boat, which is about 20 feet on the waterline, and can be pushed to just over 6 knots. If you want to break this hull-speed barrier, get a multihull with waterline beam less than about a twelfth of lwl.

While you are correct that the intercept is rubbish, if you confine your analysis to the range that the data covers you will find that the two coefficients do explain the variation in speeds pretty accurately. LWL is an indicator of the speed capability and SA/D is a measure of power to exploit that capability. So the changes in speed (and therefore potential passage times) resulting from changes in those two variables can be predicted as explained in the second paragraph.

However, this has little to do with the all encompassing terms "Sails well" or "Good sailing performance", which attempt to capture something far more complex than just speed potential.

 

[estarzinger]

I don't buy it I'm afraid. Unless I've misunderstood they did a simple linear regression. This is obviously inappropriate as the speed is zero bounded, and also displacement hull speed is known to be proportional to the square root of the waterline length - which is pretty good hint that you're not going to get away with pretending the relationship is linear. Having done a linear regression on non-linear data they end up with an intercept of 2.62, which is completely meaningless, but they make a half-assed attempt to explain it. In all, the passage you quote looks to me like complete statistical twaddle, and I'm surprised that it is published.

The maximum displacement hull speed of any hull in knots is roughly equal to the square root of the waterline length in feet times1.34 or thereabouts. (With the exceptions of very narrow or wave-piercing hulls, and planing hulls.) This means that a lwl of 20 feet can just be pushed at a MAXIMUM of six knots. (OK, if you apply an absurd amount of energy you might be able to make a non-planing hull plane, but something would probably break.) The hydrodynamics of why this is the case are well understood. I plan on 4 knots for my boat, which is about 20 feet on the waterline, and can be pushed to just over 6 knots. If you want to break this hull-speed barrier, get a multihull with waterline beam less than about a twelfth of lwl.

You misunderstand what we were trying to do. We were NOT trying to develop a physics theory of passage speed for all boats, which would be very complex and obviously non-linear. We aimed to develop a simple formula that had good predictive power for the conventional size range of cruising boats. The one we published met that aim, being both relatively simple and also relatively accurate.

We in fact looked first at the sqrt of lwl, and it had less predictive power.

Regarding multi-hulls . . . Two incidents caused us to write that article, one was several new monohull reviews for boats 40-45' that said something like "easy 200 mile days" when we knew that was total bs, and the second was a gunboat 66 passage that ended up being an 8kt average . . . . Slower than most people would have expected. Our ending formula, while not aimed at multi's actually does a decent job predicting their passage speeds.

As an FYI . . . I have a maths PhDs, and my first career was as an insurance actuary. I am in fact familiar with the basics of statistical analysis.

 

[Amulet]

You misunderstand what we were trying to do. We were NOT trying to develop a physics theory of passage speed for all boats, which would be very complex and obviously non-linear. We aimed to develop a simple formula that had good predictive power for the conventional size range of cruising boats. The one we published met that aim, being both relatively simple and also relatively accurate.

We in fact looked first at the sqrt of lwl, and it had less predictive power.

Regarding multi-hulls . . . Two incidents caused us to write that article, one was several new monohull reviews for boats 40-45' that said something like "easy 200 mile days" when we knew that was total bs, and the second was a gunboat 66 passage that ended up being an 8kt average . . . . Slower than most people would have expected. Our ending formula, while not aimed at multi's actually does a decent job predicting their passage speeds.

As an FYI . . . I have a maths PhDs, and my first career was as an insurance actuary. I am in fact familiar with the basics of statistical analysis.

It is true that the linear regression will give you the best LINEAR fit for non-linear data. Without attempting to model the entire aero/hydrodynamic problem it'd be quite possible to transform the data on the basis of known non-iinearity (zero bounding for example) to give a more defensible analysis, and if it doesn't improve predicability then you ought to be worried about your method.

You could argue that all observed data lie within section of the parameter space where a linear model gives fair prediction (ugly - I wouldn't publish it without some justification of the fudge). Given that the model is wrong, I'd want some measure of significance of the findings, and given the fact that we're applying a linear model to non linear data, I'd not believe anything about the intercept.

So I still don't buy it, but won't battle on if you say it is a good enough. I still don't buy it, but I guess that's my problem.

I meant absolutely no criticism for not including multihull data in the model - you wouldn't the same model to work. All I was saying is that, in cruising yachts, if you want to break through the normal "hull speed" barrier you probably have to get a multihull. Implied but not stated was that, if you are asking to get six knot passages from a Contessa (lwl about 20 feet), you are asking to drive it at hull speed continuously. That is clearly impossible in normal variable sailing conditions and so every bit less than six knots has to be compensated for by speeds over six knots, which is to all intents impossible, so whatever your detailed analysis, you can forget the idea of six knot passages in a Contessa.

We should agree to differ on this rather than subjecting the forum to a tedious battle, so you can have the last word (I don't have a maths PhD).

 

[Poignard]

Maybe 'good sailing performance' is the yachtbroker's equivalent to the dodgy used-car dealer's 'drives well'.

 

[stav]

I think I might have been through all the ponderings listed here, vis-a-vie boat,family,cost,wife. I have decided to keep the conway I have and will singlehand her and at least the family will still come away for a couple of weekends and a few days in the bays. For me, whilst she is not a fast boat, fair passage times can be made and also if the weather turns she has a big enough engine to get me home. When looking to down size I came to the conclusion that the smaller I went the more modern the boat needed to be as the thought of 4knt passage times seems too painful. Years ago there was an article on a chap who had cruised a 2 1/2 ton hillyard for most of his life and I think the average speed was 2 or 3 knots.

My list included hustler 30, trapper 28, twister etc. Cost was more the issue to save the money and so to find the best value for money boat. I nearly bought my father's contessa 28 but I still have the hankering for offshore passage so would have gone for the hustler/twister option.

He does have the 28 still for sale but but you would need to get rid of the smell of dogs, good price though.......

Good luck with it.

 

[onesea]

Maybe 'good sailing performance' is the yachtbroker's equivalent to the dodgy used-car dealer's 'drives well'.

SeaJet & Twister_Ken
I am not knocking what you are saying I would class short tacking as sailing hard, particularly a 43' footer of any age. Although I do think its a valid point depending on the OP's sailing grounds. I had made it clear I was talking generalities rather than specifics/ exceptions I was also thinking more in the mid 20 foot range.

As for this formula have I got it right:
D2= Sail Area Sqr Foot
D3= Displacement Lbs
D4= Waterline Length Feet
Knots= 2.62+0.066*(D2/D3)+0.051*D4

 

[30boat]

On a windy passage to Madeira my Fulmar averaged 5.8 knots.That was in reasonably fresh conditions and on a reach.Coming back beating all the way I got a little under 5 knots.I guess these are the realistic speeds for a boat of this size and age.A modern wide hull would have been quicker reaching but probably slower beating.

 

[estarzinger]

As for this formula have I got it right:
D2= Sail Area Sqr Foot
D3= Displacement Lbs
D4= Waterline Length Feet
Knots= 2.62+0.066*(D2/D3)+0.051*D4

Negative . . . The "sail area to displacement ratio" is not SA/d . . . . The correct formula for that ratio is shown later in the article. There is a 2/3 power in it. SA÷(D/64)^ 2/3

This is one reason the "critique" of the formula above is off base. Amulet keeps saying the formula is "linear" when in fact it is not. We looked at various non-linear factors, and picked the simplest best fit model, which in fact is "non-linear" at least in displacement. The other problem with Amulet's critique is that he keeps focusing on the lwl zero intercept and that it is not zero, but the physics of a near zero length vessel will be quite different than of a "normal length" one, and so like with a multi-hull, a different model would be needed. One model will never be able to cover the entire range of possible sizes and designs (think of kite boards!). I might also note that this formula is clearly (and several times) referred to in the article as a "back of the envelope calculation", very clearly communicating that it is intended to be a relatively simple way to get a relatively accurate feel for passage speed rather than anything more than that. It did explain 95% of the speed variation in both our passage data set and in the IMS fleet vpp.

 

[onesea]

Negative . . . The "sail area to displacement ratio" is not SA/d . . . . The correct formula for that ratio is shown later in the article. There is a 2/3 power in it. SA÷(D/64)^ 2/3

D2= Sail Area Sqr Foot
D3= Displacement Lbs
D4= Waterline Length Feet

=2.62+(0.066*(D2/(POWER((D3/64),(2/3)))+0.051*D4))

Have I got it right this time?

 

[estarzinger]

^^ it now looks right, but I don't have the article in front of me at the moment. I would need to copy it into excel and check it to be absolutely sure.

 

[LittleSister]

D2= Sail Area Sqr Foot
D3= Displacement Lbs
D4= Waterline Length Feet

=2.62+(0.066*(D2/(POWER((D3/64),(2/3)))+0.051*D4))

Have I got it right this time?

If you go to estaringer's website (see his/her signature for address) you can download an excel spreadsheet that'll do all the calc's for you - you just enter your own boat data.

(This is just one of many interesting and useful things on the site - Thanks Beth and Evans.)