A better "mouse trap" is available : rating system

Here I provide an implementation of the NMBR system.

I write AN implementation as I've build in a few fine-tune dials. These can be tweaked in order to have the system reflect the reality better.

I've performed an initial tweaking on the data/experience that I had. However it should be regarded as a starting point.

It is not possible to tweak an individual design, I'm strongly opposed to that and think it is direct cause for abuse. All tweaking impacts on all cats. It impacts on some more than others, just like physics would work in real life. This is in my opinion a safeguard against abuse and unfounded tweaking. The way it needs to be done now is that a phenomenon that gives cause to differences must be identified first and by founded by physics. After which it can be implemented or tweaked and all boats get a proportional correction dependent on how much they were affected in reality.

Again I stress this NMBR system is not perfect and I myself can identify some outlying points that look a little of the mark. HOWEVER, I do believe these outliers to be a lot smaller than those that are present in the alternative Texel/ISAF and yardstick ratings.

As of yet I have identified two issues of concern :

The comparison between double handers and singlehanders when a spi is added. I have a feeling that here a phenomenon is at work that limits the performance of singlehanded spinnaker boats that has not be incorporated yet. They all seem to underperform to their ratings. The spi-less singlehanders seem to be alright but the spi singlehanders aren't. One easy solution would be to reduce the spi hit that these boats get and thus make is less than the spi hit on doublehanders. This has however not been implemented yet as I have very little race data that compares the spi equipped and non-spi versions of the same boats. I'm hoping that my current contacts with the US raters can help me here. Anyways, research is continuing here.

The second issue is that to get a ready to go system I've made use of the Texel formula as the working core. I use it less than the orgininal Texel system and that allows me to take out a few defined issues. However there is one issue that I feel is fundamentally wrong with the Texel formula. I truly believe that it makes an error with the M20, M18, A-cat, F18HT and other lightweight boats that are LONG for their displacement. These boats have cut down on wave-making drag in an enormous way but hardly on the wetted surface area. Of course in light winds there is hardly any wave-making drag. Below theoretical hull speed wave-making drag can be pretty much neglected. However Texel still hits longer boats with several points.

There is one experience that I can't tweak into the system as it is. Glenn Ashby has mentioned in the past that he felt that the Taipan 4.9 in the light stuff was very comparable to the modern A-cats. And when looking at low speed drag (wetted surface dominant) such an experience is supported by physics. However it is not by the Texel/ISAF formula's. For the sceptics under us I have confidential data where an World Top-5 Tornado crew on a sloop Taipan raced several races against a World top-5 A-catter on his modern A-cat in the light stuff (around 5 knots). The Taipan crew won all races except 1; there were some 5 races in total. The physical model I derived by other means completely supports such an outcome; Texel and ISAF core rating formula does not. The thing was that wetted surface drag was completely proportional to the other factors and sail drive thus making the drive/drag ratio of both boats very comparable. On this data and others I think texel and ISAF may be off by 3 % in rating the M20, M18, A-cat, F18HT and other lightweight and long boats. (on decreasing order)

Best thing to do would be to regress a new rating formula and replace the formula derived from Texel by that one. I think that would solve the final issue with the NMBR system.

Again, by using the Texel formula the NMBR can never be worse then TExel or ISAF only better as it is corrected for other issues.

What is incorporated in the given version of NMBR ?

-1- Boats get a larger hit by adding a spinnaker.

The 4 point hit was rediculously low especially to the cat-rigged boats. It felt wrong to the sailors as well when looking at the speed gains on the water. The endresults and errors were masked by other "corrections" so the the Texel system did alright in most situations. Mostly by luck and coincidence. One of these was that the jib hit kept on counting when a spi was added making the ratio between sloop spi baots and sloop non-spi boat correct. Of course in relation to cat-rigged boat the whole system went completely beserk.

-2- Sloop rig boat adding a spi gets a reduced jib hit

Simple reason : because that is what happens on the water and in theory. The gains of a jib are lost on the downwind legs when a spi is set and takes over the workings of the jib. We are left with only a smaller gain on the upwind due to having a jib. This phenomenon is now reflected in the rating system

-3- Speeding up or slowing down of boats with increasing winds is incorporated

Simply because this happens. Great example is the US I-20 to a Hobie 16. In the light stuff the difference is alot bigger than a 20 knots. We all know that and we have all seen it. Similar things are reported by Dart 18 sailors when comparing themselfs to H16's. A similar thing is encountered between the I-17 and I-17R as well as between the FX-one and I-17. For this we have abundant anecdotal evidence and race data. Brobu on the virgin Island has commented on this often on this forum

-4- Removing a jib impacts more on a spi-less boat then on a spi boat and in the last case the sloop boat is only a little faster around the cans. (When large reaches are included things may be different. Nut that is distance racing)

Thanks to Tornado Alive I had a good key to tweak the system to. By accident he and his trainings buddies ones races together where one of the boat had to sail without a jib. Because they were trainings buddies we knew how the crew skilled compared. Downwind = No difference, Upwind : in the puff the VMG was comparable. In the lulls the cat-rigged tornado fell slightly back and had to drop a man from the trapeze while the sloop tornado crew could stay out and continue. This suggest that maximum righting ratio (the same in both cases) kept the boats fully powered up tot the same drive as long as the wind was strong enough. When it wasn;t the extra drive of the sloop rig gave the sloop an advantage dispite having to sail a little lower. At the bottom of the excel sheet ( or picture of it) you see this example in rating numbers. By virtue of general rules impacting on all other baots this phenomemon has been extrapolated to all other boats. This solves a big issue that was present in both Texel and ISAF and does closely approximate what a Yardstick will show after several years of gethering data and converging (if ever gethering enough data)

-5- Boards, skegs and asymmetric hulls are all hit differently now

Boards are better then skegs, skegs are better than asymmetric hulls. I've decided against including a formula for boards efficiency in this system like ISAF does for two reasons. I think the formula used is inaccurate. I base this on experiences expressed to me by the Stealth designer. Secondly; it assumes that the hull itself is relatively unimportant. In case of the Stealth we had some good data that said that smaller and low aspected boards only tended to be disadvantaged in the medium wind range. The Stealth design went through 3 different board setup before settling on one setup with the F16 version. The funny thing is that sails tend to show a similar behaviour. Flat in strong winds, fuller in medium winds andback to relatively flat again in the light stuff. It appears boards behave similar. The ISAF formula does some funny stuff by looking only at aspect ratio and not at size. All in all ; I felt that the used formula introduced more errors then it provided increases accuracy. When noting that measuring boards is another thing to do I decided against a complex board rule and keep only adjust Texel where we could do so easily. What we did was keep the boards hit (we can increase or decrease that when necessary) and gave the outdated asymm hulled boats an extra deduction in speed (1 point) when they satisfy the "3 out of 5" rule

-6- 3 out of 5 rule; old boat society

All cats satisfying at least 3 of the following conditions gets a rating one point slower
-1- Cat has assymetrical hulls without boards or skegs
-2- Cat has a mast unsupported by diamond wires or extra shrouds
-3- Cat has a pinhead mainsail
-4- Cat doesn't have a downhaul system nor a mast rotation system that can be continiously adjusted
-5- Cat doesn't have a smooth (rounded) transition from its sides to the deck (deck lips as on H16 and P16)

This ends the abritrary good will points that we included in the Texel system and improves on ISAF that doesn't recognize that these boats are less efficient in design as cat designing has progressed.

-7- Oversized spi hit.

1 point faster if your spi is oversized. Definition according to Texel and ISAF system. It is to much to explain why the Texel spi rule works. It was a lucky guess when it was implemented but it seems to work well. If you do math on it you'll find that it forces the spi's to a constant ratio to the boat size. It keeps drive and drag ratio constant. However we have oversized spis on some boats. We can't forbid them. So we enter the simple rule that 1; no designer will fit an oversized spi when he thinks it is slower. 2; more than enough data suggests that oversized spis are really limited in gains (Nacra NE vs US I-20 anyone ; Tornado vs EU I-20 ?). So 1 points seems to be the minimum (36 sec on hour) and also seems like a good maximum. It sure does simplify things alot. ISAF formula for spis doesn't seem to discriminate much.

-8- Singlehanders are relatively faster in light airs then in heavy stuff

Self explanatory; On of the major bitch points about using Texel and ISAF. Now resolve in a simpler way then Yardsticks do (namely US PN with 5 groups)

-9- Two rating numbers assigned to easily recognisable conditions

Race committees no longer need to bring a windgauge on board and do repeated measurement. Big bitching point of the RC's; at least over here. Now the RC can look out of the boat and estimate how many boats are fully trapezing when going upwind. More than 60 % => high wind rating. Less then 60 % => light wind rating. Sure we can still find border cases where an offset can be found but these are MUCH smaller than Texel and ISAF and Yardsticks groups are either not used often or have insufficent data for all groups to be dependable. Especially when rating new or rare boats. It is also alot simpler to use and each boat gets a correction proportional to how much its suffers. Not perfect but close enough

-10- Some things but it is already late and I'll let you chew on this stuff to begin with

Good luck !

Look in the next post for the excel data sheet , I removed it to make reading the above post easier

Wouter

..

Hi Wouter,

Take your time - it's worth it.

Let me know if I can be of any assistance.

Here's a super simple formula I derived for time-on-time scoring:

TOT Handicap = .0673 x (SA^.584) x (DISPL^-.325) x (L^.624)

where SA = total sail area in square feet
Displ = displacement in pounds
L = length in feet
the ^ means raised to the power

Sorry, no width consideration - but this was just a quickie attempt. (that seems to work rather well, though).

If you invert the TOT handicap and divide by 100, you'll get numbers very close to the USSA D-PN's for most boats (in non-spin config). Try it!

Take care and good luck!

Steve

Wouter,

Quick questions

1, What identifies an over size Spi - is this the same as the Texel definitions ?
2, Would you mind sending me a copy of this to scooby_simon (at sign) ntlworld.com so I can have a play with it later and print it so I can understand it properly. The excel will not print in any way that will be readable from here.

Steven,

This exactly what I mean !

Many sailors think yardsticks like US PN adjust each rating to little quirks in each design, but it is actually not alot more than a large table of numbers that follow a very simple formula.

This is exactly the reason why Texel and US PN give very comparable results despite appearing to be very different. Of course a few exceptions exist.

If we know this then why not make it easier on ourselfs and use this knowlegde to make the system alot simplier and more transparent. Arguably the benefit of a Yardstick is largely a believe that it could be better even though in reality it may not be much different from a measurement system.

Thank you for your support, I need it and it seems others are convinced by it that this is something we cat sailors need to get of the ground. I've been contacted by some already, some even with influence. Hopefully we can get this of the ground.

I will let you know when you can be of assistance. Actually you can already. You can spread the word around in your local area of cat sailors. We need sailor support to convince the power that be that something can be done about many complaints that sailors have of the current setups.

If we all together don't ask and demand such a thing than we will never get it.

Sure they will ry to shoot this done but as our moto says; It may not be perfect, but is may well be better !

So lets go !

Wouter

Speaking of Jibs and Spinnakers.
How about the Hurricane 5.9. They can put on a Spinnaker, use a slightly smaller jib and stay at the same SCHRS Number

Quote

Speaking of Jibs and Spinnakers.
How about the Hurricane 5.9. They can put on a Spinnaker, use a slightly smaller jib and stay at the same SCHRS Number

But adding a little weight too.

The Hurrricane 5.9 jib was always too big anyway.

Scooby,

-1- Same as Texel definitions (for now) meaning

up to 16 foot => bigger than 17 sq. mtr.
16 to 18 foot => bigger then 21 sq. mtr.
18 to 22 foot => bigger then 24 sq.mtr.

I'm considering two small modifications, but they are something for a later stage. For now this works well enough. To give you an idea. Have different limits for singlehanders. Currently an A-cat can have 21 sq.mtr. and that is a bit much.

-2- I don't mind, everybody and anybody wanting to see the excel sheet can have it as long as they agree that it is my intellectual property !

It is in your inbox, Scoob !

Wouter

A cat with a 21 kite would be (very) silly

Got the sheet

Thanks....

Quickly spotted one slight error, the weigh od the Inter 17 is 161 with the Spi rigged (and it makes a difference of one point on rating)

Also, I dont really think an Inter 17R(spi) is quicker than an F18 in light and the same in heavy.

But it is a start

Good work

How did my Hobie 20 post jump to Wouter's "Mouse Trap" thread It is listed in both

I'm think that may actually be acurate...or close to it. The I17R carries less crew weight and points higher.

But also less righting moment and less pairs of hands

Which rating sys will bring the SC20 rating down to a realistic level? That's the one I'm for.
JC

-1- I used the Texel core so all weights are excluding the spi gear, take a look at the F18, F20, F16 weights as well It is something I want to take out of the system as well as stated in the "Potential other mods" post, but I haven't done it yet. You can pretty much 1 point to all spi boat ratings if you use the weights including spi gear and then subtract 1 point because I will lower the spi hit to compensate. End result ? Zilch. That is one reason why I haven't implemented it yet. I'm waiting on some feedback from others.

-2- I-17R, this is an identified problem point, see the post publicizing the numbers. I'm hoping that US data and Mark Schneider can help me here. I don't have good I-17R with spi data. There is one thing to say for the NMBR rating for the I17R. This boat has almost the same mainsail and mast as the F18's on a lighter platform with less crewweight and a considerable spi. If the fundamental Texel formula is right (Not entirely sure here see post) than the I-17R should be faster in light air ! However I think NMBR makes it too much faster. I feel the cause for this is found in the Texel rating formula itself. Hence my suggestion to do a new regression on the data. In summary; we are working on it.

Also please correct the wrong width for the I-17R ; the sheet says 2.60 mtr. but it should be 2.50 mtr I believe.

>But it is a start,

I would say look at the other things, these seem to work pretty well. Of course it is a bit much to ask to get it right the first time. Some tweaking is required.

Wouter

I(F)17R should be 2.5 wide (same as I17)

F17 press release

Hey Jake,

You can help me here. (either via the forum or in private)

I don't have much data on these singlehanders and I'll need that in order to tweak the single handers. You as a F18 sailors may help us out here.

How often have you sailed against these singlehanders and how did they compare to your F18 ?

I'm particulary interested in the I-17R as that is such a clear example and I can cross reference it with I-17 data. These are great setups; Identical setups with only a small of differences. Makes for great tweak keys.

Wouter

How about a NMB rating for the SC20 TR with spi of

Light weather 88
heavy weather 92

SC20 TR no spi :

light 91
heavy 96

Compare to US I-20

Light 92
heavy 95

Seems a whole lot better doesn't it ?

I used the following specs (please give me the correct one if they are wrong)

weight 200 kg (441 lbs)
length 6.10 mtr (20 foot)
width 3 mtr (10 foot)
Mast 11.5 mtr (38 feet)
Mainsail 22 sq. mtr. (236 sq. ft.)
Jib 6 sq. mtr. (65 sq. ft.)

And I only needed 60 seconds to acquire a rating and that included looking up estimates of the specs via internet

Is that satisfactory ?

Wouter

>>But also less righting moment and less pairs of hands

Wait a minute. You can't use the righting moment excusse here ! In light winds (NO TRAPEZING group) righting moment is not an issue as all crews can then generate more rigting moment then they need.

And in the strong wind rating the righting moment was calculated and used to compensate the rating. The I-17R received a swing of 6 rating points (3 minutes and 36 seconds) while the F18 good a swing of nothing ! By this the I-17R is rated slower than the F18 in the heavy weather rating (trapezing group)

You may think the swing is still not enough but you can't say that NMBR doesn't take "less righting moment" into account because it does.

>> less pair of hands.

Well yes, that is another matter although in the light stuff this may not be much more of a factor than 1 or 2 rating points. I mean how much longer do you need as a singlehander to set or take down a kite in race. A few second. Will 36 to 72 second compensation be good enough for that ?

By any account this is only a small offset.

In strong winds I say "less pair of hands" is a bigger factor.

Again guys please look at the other things as well. The single hander with spi was already identified as an point of attention in the post giving the NMBR table. The spi hit to singlehanders seems to be a bit harsh, I'm contemplating what to do with it already.

What do you guys think of the H16's versus F18 for example or the evert recurring Curacau US I-20 to EU I-20 issue ?

Wouter

Hey Wout,

The SC20 is 12' wide, but i'm not sure about the wt and sa. What does the change in beam do to the #?

Here is the rating correction for the 12 foot wide SC20 over the earlier presented 10 feet wide numbers

New specs

weight 200 kg (441 lbs)
length 6.10 mtr (20 foot)
width 3.66 mtr (12 foot) (This is some mightly wide boat)
Mast 11.5 mtr (38 feet)
Mainsail 22 sq. mtr. (236 sq. ft.) luff being 11 mtr
Jib 6 sq. mtr. (65 sq. ft.) luff being 5.5 mtr.
+ spi

New ratings SC20 at 3.66 mtr width

Light weather 88
heavy weather 89

Compare this to :

SC20 at 3.05 mtr width

Light weather 88
heavy weather 92

And the US I-20

Light 92
heavy 95

If we take the spi off the SC20 than :

Light 91
heavy 93

I think this reflects the (scarce) real life data well. At least heaps better then PN of 64.1 vs that of 59.2 to the US I-20

Notice how the new width didn't change any off the light weather ratings (No trapezing group); only the strong wind ratings (Trapezing group).

Again we needed only a few seconds to get the new numbers.

Also interesting is to compare the "sc20 at 12 feet width" numbers tot the ones of the M20

M20

Light 86
heavy 90

So the M20 is expected to be 2 points faster in the light stuff and a fraction slower (1 point) in the heavy stuff.
The M20 has a 4 point rating swing from light to strong and the SC20 only 1 point. This all due to the ratio between rig and width. The Hobie 16 (L = 116 ; H = 114) of course has a opposite swing (getting faster) of 2 points. So the difference between the M20 and the H16 reduces from 30 points to 24 points when the wind conditions transitions from no trapezing conditions (light winds) to Trapezing conditions (heavy winds). The difference between the SC20 and the H16 only swings from 28 points to 25 points. To go one further US I-20 (L92-H95) vs H16 : difference goes from 24 points to 19 points. However the difference between the H16 and a nacra 5.0 remains about the same (0 point differnce to 1 point difference). Both of these speed up equally due to changing wind conditions.

Isn't this reflecting what we see on the race course ?

Wouter

Hi Wouter

I have been rescoring our buoy's regattas with the new rating scheme and the results look pretty good in that the elapsed time differences needed to win are pretty close for what I would argue are sailors of equivalent ability.

I had to guess at numbers for the Hobie 20 and the Shark is there any data in texel that can be used to compute a pair of ratings.
Mark

Thanks alot Mark for your first report on running the numbers on your race data.

It makes me feel very good that the initial test shows some encouraging performance.

I'm looking forward to the report after you have run it all. You know some tweaking can still be done.

You what's funny, there are some points I think to be remarkable but it may well be that my own prejudice is unwilling to believe what the formula's are saying. So again, I'm looking forward to your commment and or report after you've run the data on it.

Texel data :

http://www.texelrating.knwv.nl/

Hobie 20 :

Best look up the Miracle 6.0 in the Texel rating and use that data. It is my believe that this boat is the same as the US Hobie 20 Miracle. One point of note though, Hobie NEVER officially imported the H20 Miracle to Europe. So these are "privately" imported boats and may have undergo some modifications with respect to the US versions. Never the less it is a good place to start.

Shark :

There is a shark catamaran in the listing but I'm sure wether that is the same Shark that you use in the USA. ONly way to know for sure is to contact the Shark class association and ask them about the specs.

I can tell you however that IF the Texel listing doesn't have the numbers on a boat that NO other rating system has them. Not ISAF and certainly not the Yardsticks. The Texel listing is by far the most extensive data base on catamarans around.

If a boat is still not named in the listing that the following rule of thumb works pretty well.

Length : Take the length overal or waterline length. The rating is not very dependent on which one you use.
weight : Often this is easy to guess. 14 ft = often about 100 kg, 16 ft = often about 135 kg, 18 ft = about 160 kg 20 ft = about 190 kg
Sail area = Take an F18 sailarea and scale this area in height to the mast area and in width to the hull length.
Luff length main = mastlength - 0.5 mtr. (a very good estimate)
Luff jib = nearly always between 4.25 and 5.5 mtrs. depending on the overal size of the boat. A rought guess will do here

Wouter

mark,

This is what I found after some 60 seconds on the internet and thanks to a post by Same Evans.

http://www.sharkcatamaranclass.org./WHAT.HTML

Shark specs

Length 20 ft (= 6.10 mtr)
Width 10 ft ( = 3.05 mtr)
Weight 450 lbs (= 204 kg)
Mast Height 28.5'= 8.69 mtr => give mainsail luff estimate of 8.19 mtr.
Sailarea total 275 sq ft. = 25.6 sq. mtr. => lets divide that into 18.6 sq.mtr. main and 7 sq.mtr jib
Jib luff should be about 5 mtr with a mastlength and boat width like this.
daggerBoards
No spi

These estimates (and facts) give the shark a provisional rating (we had to guess at some specs) off

Light 99
Heavy 98

I ran the numbers on the Hobei 20 as well (the number that I know or estimate)

specs

Length waterline = 5.82 mtr.
width = 2.5 mtr (?)
weight = 201 kg
main = 17.97 sq. mtr. by 8.81 mtr luff
jib - 6.34 sq. mtr. by 5.74 mtr
Boards
No spi

NMBR ratings

light 100
heavy 102

Mark I send you an excelsheet so you can punch in the numbers of some other boats you need yourself

Wouter

I appears that the Texel rating listed Shark is about the same boat. Only texel lists a shorter mast and less overall sailarea.

On these texel specs the shark get the following NMBR numbers

light 101
heavy 99

If we have to guess at it a number around 99 seems to be right. Only this design is pretty old and arguable less efficient. I'm not to sure wether the NMBR rating gives a very accurate prediction here. Afterall NMBR estimates potential for a given box of specs. It is up to the skills of the designer to bring out that potential in real life. ONE-Design classes of 40 years are arguably at a disadvantage here. They can grow with the new developments.

But I anxious to hear how these numbers come out of your tests

Wouter

The Shark was originally designed to fit into the IYRU B-Class specifications, with 235 square feet of sail. But the Sharks that were exported to the United States (and those that were subsequently built in North America) had the higher sail area of 275 square feet because the United States, in general, has lighter wind conditions than Europe.

So any Sharks sailing in Europe may have the original sail area of 235 square feet. (I don't recall ever hearing anything about a difference in mast height between the Sharks in Europe and those in the U.S.)

Plus, the Shark has centerboards, rather than daggerboards, and the maximum depth is 3'6".

NMBR is easily adjusted to produce distance racing handicaps.

The idea behind it is quite simple and yet simply copying the same approach to Texel / ISAF and Yadstick systems will not result in the same accuracy.

Here an explanation of how it works

Pretty much we can devide any race over three main courses.

-1- Pure upwind sailing (as high as one can)
-2- Pure down wind sailing (as low as one can)
-3- And the widened area of reaching (anything between the regions of -1- and -2-)

-1- is most impacted by righting moment and pointing ability.
-3- is most impacted by having a spinaker or a jib
-2- is mostly impacted by good hull design and efficiency of the traditional sailarea (only main and jib)

Right at this moment both the spi hit and the righting moment correction are implemented as follows. (I only give part of the equation, only the working core.

(0.5 * spi hit + 0.5 * righting moment corr) / 1

The 0.5 is actually a course distribution factor. We have currently taken 0.5 as most data I have right now suggest that a non-spi boat takes about roughly 50% of the time to go upwind. Some tweaking can be done here.

By changing this course ratio to say 100 % upwind and 0 % downwind we can produce custom handicap numbers for races like the Tybee 500 and steeple chase. Of course any ratio between 0 % upwind work and 100 % upwind work is possible. By changing these ratio's one can see that the impact of both the spi hit and limited righting moment hit are forced to influence the ratings proportionally to the sailed course.

Smart people will notice how the reaching leg in not implemented yet but by adjusting this formula too :

(Downwind portion * spi hit + (1 - downwind portion - upwind portion) + upwind portion * righting moment corr) / 1

and use 2 percentages to defined the course. Portion upwind and portion downwind. The reaching portion is then the remainder.

This way we can modify the framework by including the reaching legs. On these legs no spi hit or righting moment corrections are taken and so the pure speed potential under traditional sails (main and jib) is taken and assumed to be fully powered up all the way over the conditions.

The framework is simple to implement and by adjusting just two input variables the excel sheet will produce ALL the new custom handicaps for a distance race within a blink of an eye.

It can be proven theoretically that the accuracy of the rating is somewhat less than those for a well layed out bouy race (for which the rating were optimized) HOWEVER, the same proof shows that the NMBR custom rating will be noticeably more accurate than all other systems in use today. Simply because it DOES compensate for the different course shape of the distance race AND because it does so in a way that is SIMILAR to what happens in real life. To how real life physical processes determine the performances under the different distance racing conditions. Ex. We all understand how sailing 90 % of the time upwind and 10 % of the time downwind under spinnaker proportionally impacts on the benefit of having a spi. The shown framework does compensate in exactly the same way.

The same framework can not easily be copied and used with Yardstick systems as in Yardstick systems the distribution between upwind, downwind and reaching performance is simply not known. One needs this in order to have them impact on changing course distributions.

Texel nor ISAF can easily use this setup untill they change the way that they implement the spi hit. By this I mean that they can't use it untill they have decoupled the spinnaker and jib related performance on the downwind legs. This is pretty fundamental stuff and many important people unquestionably want to discuss such a thing at great length.

So distance race organisers !

There is also a better mouse trap for you guys.

Wouter

The plan to get the NMBR system accepted.

The first conclusions by some reviewers (independent from my person) are that the NMBR system, in its current stage, appears to produce more balanced scores and put crews of assumed equal ability closer to together. One comment was that it actually looks pretty good. I'm quite contend with that.

But this puts us to the next level.

There are two things that need to be done :

-1- additional testing on real life race results and possible tweak the system a little here and there for extra accuracy

-2- Start up a program to contact all the rating committees and get the system accepted.

Point 1 and point 2 can be executed simultaniously as more and more signs are coming in that NMBR is producing better results than its alternatives. So additional accuracy is welcomed but not necessary. Also the largest gains of NMBR are to be found in its ease of use. It involves much less work than a yardstick system. For the ISAF/Texel corner the NMBR system solves a few claring issues and produces much more realistic results for the same effort in maintaining it as the ISAF/Texel systems currently in used. Also it is more flexible as it can produce custom ratings for one-way handicapped races etc.

Right now, I'm working out a plan to tweak the system further; I'm in contact with a US party for that.

With regard to acceptance I propose to go about it in the following way.

-1- Consolidate contacts with the USPN committee. Right now they are the party to most benefit from it and it will allow them to offer a remedy to RC considering breaking away from USPN in favour of Texel like rumour has that Tybee 500 is considering. In short the time appears to be right to expose the USPN to an alternative.

-2- Next step will be to contact ISAF as I know they are looking to improof on their system for several years now. Not much is happening so I think they are at a dead lock. The medicine might well be to expose them to a fully operational system that US parties have tested for improved accuracy. Simply put we can use the US to break open the possible deadlock inside the ISAF committee. Afterall, ISAF would like to see a single system around the world and the US is very much an sizeable and important block in that. Also this is the most effective way for US sailors to influence ISAF into a system with wind dependent handicaps etc. In short to make them accept some of your wishes/demands. Of course ISAF is looking to stuff Texel as the more dominant system. So the US is the juicy bone and having a noticeably better system is the rewards for ALL OF US.

-3- Of course if USPN and ISAF go like domino's than we can really put some pressure on the Texel system and most likely get them to accept to NMBR system in it total without counter demands. I think I have the inroads to that already. I hope my former collegues at the Texel system will forgive me for this but the current make-up of the committee is in need of a make-over and nearly all parties cooperating with the Texel committee adhere to that view. So in short is the old committee holds out despite convincing arguments then I will contact the parties using the system directly and pry them of Texel that way. It will be their call to decide to co-opt or be made disfunct.

-4- Yardstick systems in general. Well Yarstick in the UK is already losing to ISAF at this moment. The transition is slow because of the large and known issues in ISAF. With NMBR these are as good as solved and so further resistance will be solely based on emotion and not on verifiable concerns. Asia, africa, and south america are largely to small to maintain their own independent yardstick systems. Alot of them already use ISAF or TEXEL because that is much easier for them, when the US and the EU are on the same line regarding a handicap system then surely they will all transition to that system over time as well.

-5- Australia and New Zealand. This will be tricky as these scenes tend to be very independedly minded. However right now several yardstick systems are still competing with eachother also the dominant VYC suffers from all the issues linked to yardstick systems and uses on 1 rating number for all conditions. Arguably for them a large improvement can be gained by using NMBR. The largest of all. It will probably take some massaging but also when US and EU are in agreement than such a thing must go a long way in presuading the Aussies.

As you can see ; USPN and it committee play an important role in all this. They are the first step and those are always very important. However I do plan to be in contact with several parties simultaniously.

I'm expecting a list of demands from several parties (or should I say whigs) however I will not accept any demand that are different from a general wishes regarding the use, shape or accuracy of the NMBR system.

Simple reason being that I see no point in sacrificing ease of use and accuracy simply because a whig wants to boast that he influences the development of the NMBR system.

I will give an example ; a party can demand that a rating should be adjusted when a carbon mast is used. I for one do not claim to know how strongly and in what way a carbon mast affects performance. Nor do I know of a easy and simply way to implement a fair correction. In addition I can't even proof that having a carbon mast impacts on a boats performance in such a significant way that it requires a handicap rating correction. To give an example ; the case for hitting wingmastswith prebend and spreaders when compared to a spreaderless teardrop shape mastsis ALOT stronger and yet NO-ONE ever felt the need to hit wingmast individually. I found that the mere mentioning of carbon makes alot of sailor skittish. This is an emotional response and in many cases it is not well founded in physical phenomena or even real life.

I do very much expect from the official that they make a profound effort to understand why certain things (factors) were included and others were not.

In general I compare this to the following example.

Double glassing.

Going from single glassing to double glassing for windows gives a 50 % reduction in loss of warmth and thus leads to fuel efficiency in the home and savings despite the extra expense linked to double glassing. We can repeat the trick of course a good for triple glassing or event quadriple glassing. The gains will be additional 17 % and 8 % respectively when going from one to another. Theoretically speaking additional gains are made however in reality triple and quadriple glassing doesn't make economical sense, the costs of these outweight the potential gains. It is therefor foolish to seek such windows.

Measurement ratings systems are much the same. After a certain minimal number of inputs the potential gains of adding another input will become very limited and more often then not the additional costs of such an addition will outweigth the potential gains. Carbon mast hits were just such and example and actually ISAF has done away with those in 2003.

Wouter

Wouter,
When calculating downwind sailing, you say "as low as you can go." That means straight downwind. Is that what you mean?

Also is this the correct order of these three items? Well, I mean are the items in the right order but are the numbers on the wrong ones?

-1- is most impacted by righting moment and pointing ability.
-3- is most impacted by having a spinaker or a jib
-2- is mostly impacted by good hull design and efficiency of sailarea