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