Mike...
This rating table is the ruler we use and the issue is... how many fine gradations of the ruler does it have to measure this f18 development.
So... the data we collect is finish time on the race course.. This data has to be accurate if you want to evaluate the true performance of the boat and add more gradations to the ruler.
The simple point here to illustrate the problem is that the RRS of sailing have a huge impact on how we measure the performance of the boats because it effects the time the boat goes around the course. The measurement of the boats performance is just not accurate!
The nature of the game (RRS, wind shifts, currents, and less then perfect sailors and other factors) effects the performance times measured and is just noise.
So In essence... the ruler we use is marked off in 1/4 inch steps because of the noise in racing. ... you estimate where the performance is between two hash marks for boat classes. Now...do we need more gradations... ie should we use more decimal places to calculate finish times in a SINGLE close races?... The noise that makes it difficult to measure a performance handicap exists in every sailboat race... one design or handicap... but you can't take more samples and average the result of that skipper to eliminate these factors. The bottom line is that it does not make sense to use a ruler marked off in mils when the noise for the single race result is so great? Our 1/4 inch scaled ruler (PHRF, Portsmouth or SCHRS) does a good enough job at resolving races as best as you would expect.
Another way of saying this is... the resolution of the handicap table (PHRF, Portsmouth, SCHRS) is limited and matches the noise of a sailboat race. We have a pretty good ruler to measure with... It just is not going to resolve rat's hair differences in performance. Take it to one design if you want to eliminate the boat factor.
"a rating system of ANY KIND simply can't do any better" makes it sound like you (or someone) thinks that a handicap rule should "fix" this.
No... we can't fix this... and I am not looking for someone to fix it...
Simply recognize the limits of a handicap race. One design will do a better job of sorting out the best sailor from a sailor who differs by just a rat's hair on that specific race.
If someone were to complain that because the ratings are based on real-race data, which includes such strategic and tactical situations, they are not a true representation of a boat's speed potential; that's a different discussion, which would require a different approach to data collection for the PN system.
Well this is true... The accuracy of a performance ratings table will depend on the nature and quality of the data. (Portsmouth has no data in 2011... and PHRF is not transparent) but my take home point would be... SCHRS is pretty good as a fair accurate ruler and
very transparent to the users because it is a measurement system with a published formula. Its resolution matches the game. Searching for a better performance based ruler alternative is pushing the law of diminishing returns.
You might remember a Sailing world speed comparison test of the hot 20's back in the recent day. Supercat 22, Mystere and Hobies, Nacra 6.0's... Smyth, Danniels and others sailed the boats on timed legs and the data was published. Still it's a long way to go to create a performance based handicap system from this data. Very interesting... but not worth trying to build a handicap system from it.
The argument that started this is that the formula class development have speed improvements that are class legal... but undermine the handicap system. The fleet is not treated fairly.
My point... Data won't be available to sort this out and get a new accurate rating because the noise is too great and in the end... you can't resolve a handicap race that precisely anyway.
PHRF solutions... eg a hit for a class legal change that is different then last year's model should be .005 etc. are not transparent.. (Why .005?)
In the end... a handicap race won't be able to resolve the slight speed increase that the latest F18 development produces.. In one design racing... you will see the difference.
Before we start with PHRF fixes to impossible problems.. we should move to a transparent system first.
