A few answers that may either confuse you are come across as conflicting. However they are valuable enough to full understand.

From a measurement point of view even a strickt One-design class is a box rule. The whole idea behind a measurement based system is that of a box rule. The fact that Texel and many Yardstick converge on the same ratio's between boats suggests that the boxrule approach has merits.

With regard to the Dart Hawk, I think this design is still one of the better F18's around. Put a new updated set of sails on it and go ! I think that Dart hawk was victomized by the "bad perception virus". Its sales declined rapidly after the pounds rose sharply half way through the 90's. Then sailors assumed that the decline in sales was caused by the dart hawk being uncompetitive. Of course if sales drop you won't have as many boats in the class and then when Hobie starts buying away all the top crews than more and more people start to BELIEVE that the Dart Hawk is an uncompetitive design. Than it starts to feed on itself. I still think that the Dart Hawk is still one of the better product in F18 land. Just need to get the modern cut sails on that boat.

You refer to a difference between the Dart Hawk and nacra F18. Can you point me to data suggesting this ?

Dart 18, Dart 15 and H5.9 are simply to small as a class to act as reference classes. Furthermore only the Dart 18 has some international presence but only in Europe. Then of course the Dart 18 production will be discontinued in 2005 and it is badly situated in relation to other classes. Why use dart 18 as a reference when this design makes up less than 5 % of the fleet in all European regatta's with the possible exception of UK events. You will only get very very limited comparison data this way.

Also all the good crews are sailing in difference classes.

Also the Dart 18 design (and so to the Dart 15) are a far cry of being representative of the designs that currently dominate catamaran racing.

Last but not least; I see more merit in having a large, but less strict defined base to extract data from than a very small base of high restricted class. I can easily average out errors present in large block of data but I can't do that in very small blocks of data.

In simple word, from one perspective the named class maybe better as a reference. BUT when viewed from several perspectives the Formula 18 class as whole cleans the bank. Arguably a rating system can not be based on a single viewpoint that will make it very sensitive to errors and also very inflexible.

Sorry,

However I very much appreciate your input.

Wouter