Handicaps

[Guest]

You must be a much better sailor then.

[Deep Purple 512]

I wish

[markm 30]

Don't Force It, I think if you read the front page of the RaceTrack website you will discover precisely what it claims to be. Further, if you read the FAQ you will get a few more insights into the underlying methodologies. I'm struggling to see how you can justify an assertion that this is unpublished and unknown information when it's been available on the web for most of a decade.

 

I could of course supply a complete copy of the underlying code base and data structures but unless you're a pretty experienced database designer and procedures programmer, it won't help much.

 

On the four digit thing, well actually it goes to seven digits in the source. The reason it's four digits is that any more takes up more screen real estate on a web page without being any more useful and less creates too many ties within the rankings. Sorry if that explanation isn't as much fun as a conspiracy to make people believe it's more accurate than it is.

 

On the boat datum thing whilst that's true for some systems, no that has never been the case in RaceTrack, again, you should perhaps be more careful on facts before asserting them in a public forum. You are however correct that RaceTrack uses the linear relationship between elapsed times as it's base for comparison of boats within a race.

 

If there's information that isn't on RT that you want and that I can provide, I'll consider it.

 

If what you want is a 50 page justification of each line of code and the reasoning behind it, well, my logic goes like this.

 

RT is a boat / race history database and performance analysis tool. Knowing a magnitude more detail on how it works than is already published is a week or more of me documenting (something I truly hate) and at the end of the day, people will judge RT on what it says and how that matches with their real life experience, not on an all-encompassing document that says how it does it.

 

Some like it which is nice and quite a bit of the reason I still do it, some will think it rubbish and not look at it. That there seems to be more than one person who likes it I've always thought quite positive since there's only one person who's in the #1 ranking spot.

 

I don't think you'll find anywhere that I've claimed it is a handicap system so when you judge it, judge it on what it is, not on what it isn't.

 

My personal view is that it can be a very useful tool to aid in performance handicapping, not perfect but it can get you 90% of the way there 90% of the time.

 

Where people want to use it for handicapping, either partially or fully, I'll generally help where I can, mostly that takes the form of getting them the info they want a lot more quickly and easily than they could get it themselves. Think the Tauranga people opted for it as an option because (1), some asked and (2) if you hated PHRF and IRC you could still enter, they like entries and good on them.

 

If I had to speculate on why it does seem to be developing some popularity for handicapping, it'd be this.

 

Clubs and organisations don't like handicapping themselves, it takes lots of time and effort and they get no end of grief no matter what they do.

 

That leaves only a few alternatives (PHRF, IRC and the like) and using RT, ignoring debates on whether it is or isn't better, is in most cases cheaper.

 

Did have a small thought though DFI, we could introduce a "DFI handicap', all you need do is send me handicaps you reckon boats should be on. All the functionality already exists in RT to display results under different handicap systems so no work for me. It could be the solution to years of problems, love to help. People will see it and before you know it, owners might be demanding to sail under DFI.

 

It's actually not a jest, applies to anyone, OgreCap, Bondage Rating, surely there's one or two wanting to put their neck on the block? I'll be all disappointed now if we don't have a Bondage Rating or OgreCap come Simrad time.

[ballystick 73]

I like it that Mark is so interactive and keen.

 

After each race I look up our rating to see how well we did, I get great satisfaction out of that and I know where we went wrong if it shows a negative.

 

I emailed many people about the handicapping systems they preferred and had some good responses even from Don't Force It, it was great interaction.

 

We ran a racetrack result in Akld-Tga for all no matter what, it was a challenging race - we finished last.

 

There will never be an answer for all but if we can try our best and be happy with the successes where it is deserved then so be it....

[drbob 0]

I still reckon the 930 class needs to look at IRC, or perhaps now the new HPR, as a way to level the differences arising from the various mods.

 

Yeah, I know IRC sucks for a 930, but that's compared with other classes. Within the fleet it should be pretty good.

 

I downloaded the HPR worksheet and only came unstuck when I got to mast centre of effort or whatever it was - surely we could get somebody like BBW to work through it for us.

 

There is the issue of cost, but what's a couple of boat dollars if it avoids the wailing and gnashing of teeth... and also, all these mods and bits of carbon can't be cheap, so maybe put just a little bit of effort into the rating and we'll have a sound measure.

[Deep Purple 512]

The numbers don't support that.

 

Y88's outnumber R930's more than 2 to 1 yet.....

 

City Entries

Summer Series Y88 = 15 R930 = 13

Winter Series Y88 = 13 R930 = 13

Auckland Regatta Y88 = 6 R930 = 7

 

Time difference between first and last boats in most races is negligible from the look I had. Racetrack shows a wider spread between 88's than 930's

 

So if the Rules are so good, why is half the Y88 fleet staying away?

 

Looking on Trademe at the moment there are a heap of Y88's. A couple of really flash ones, a few well performed ones, and a heap of oddballs left out in the cold languishing unsold at half price.

[war machine 0]

Not being defensive but......

 

We will have at least 20 competitive certified one design boats at The Harken Young 88 Nationals this weekend with some of NZ's best pros mixing it up with a pretty handy group of non pro owners and crews. This is up from 18 last year and bucking the declining numbers that are being seen in many regattas around the world.

 

The Corinthian "Owner Driver Champs" in October also gets good numbers of around 16 boats.

 

Auckland Regatta was disappointing with only 6 Y88s vs 10 in 2012 but it was still great sailing and an awesome party including the relaunch of the Y88 Mt Gay "Loving Cup". Feedback was that "we ain't out of recession yet" and sailing the Friday made it too hard to get crews. This seems to align with feedback offered on other forums. Next year we may look at doing what other fleets did and make Friday an "invitation day" with Saturday and Sunday counting.

 

Managing the perfect one design rule and fleet is a big ask (almost as hard as perfect handicaps) especially for a committee of volunteers but as a fleet we continue to look for ways to improve to keep racing fair, costs manageable and protect the value of the boats as a leading cruiser racers.

 

Racing against 930s and others in the bigger events is also great fun ... perhaps we need an interfleet challenge

 

Drifting thread back on track, the Y88 fleet also runs handicap numbers to encourage those of us using older sails and crews at the middle and back of the fleet.

 

Winning on line is everything, winning on handicap is acknowledging improvement. Reference is made to Race Track when setting handicaps and it is a great ready reckoner and source of results. To the person who does all of that Race Track work we salute you.

[Guest]

Hey Mr Racetrack

 

I am just a simple yachtie and I can assure you that I am not saying "the Emperor has no clothes".

 

Some years ago when Yachting New Zealand started publishing "sail to numbers" I queried them as to how they transformed a set of elapsed times into sail to numbers (in support of their PHRF calculations). It turned out that YNZ had sourced the calculation and code from a third party and had no idea how it worked. After further investigation I discovered major flaws in the sail to number calculation and this led to some boats having significant anomalies in their PHRFs given by YNZ. Yachting New Zealand shut this discussion down as quickly as they could and the issue was not widely known of. I think those affected purchasers of the PHRF service deserved better.

 

Yacht clubs are now contemplating using Racetrack. RT's webpage and FAQs only describe in very general and vague high level terms how a set of elapsed times is transformed into sail to numbers.

 

Mr Racetrack would you be kind enough to describe in your own words specifically how this transformation calculation is made. Source code is not necessary. Please dont take this as a criticism of yourself or Racetrack in any way.

[markm 30]

Okay. Will take a day or two though, doing this in a way that's logical, at the detail level you've asked and providing context to allow it to make sense isn't something that can be ratteld off in a quick post while sipping a brew.

[drbob 0]

Late last year I had a quick look at PHRF sailed-to values because I wanted to see if I could work out how the PHRF TCF might be calculated/updated from those.

 

I couldn't determine an update formula, but I did figure that it very much looked like the sailed-to value was determined as the TCF the boat would have to have had to have finished equal first with the boat that actually finished mid fleet on corrected time (the reference boat).

 

That is, if there were 11 boats in the race, the reference boat is the boat that finished in 6th place on corrected time. So the sailed-to for the boat that placed 1st would be the TCF number that if multiplied by its elaped time would have had that boat finishing at the same corrected time as the boat in 6th.

 

I haven't looked at it lately.

 

This is the same method we use at our club to determine sailed-to for the purpose of handicap update. I can explain the formula for this handicap update if you want, but it doesn't appear PHRF works the same way.

 

I think there's a problem with any performance handicap system (so that's PHRF and RT if RT was used as a handicap system) in handicapping boats that don't sail together, when they do come together. I mean, say I usually sail at Club A and you usually sail at Club B and our fleets don't normally sail together. Handicaps will/can/should be good within each fleet. But if some boats come from Club B to race at Club A, there can be a disparity. Let's say there is an 88 at each club, and when they sail together they always cross the line within seconds of each other, yet their handicaps could be very different.

 

The PHRF difference between Pepe and Wild Oats is probably just the usual PHRF practice of starting a new boat/owner combination at the higher end of the range of TCFs for a boat of that class. After a number of representative races, ask for a review based on relative performance between the boats.

[markm 30]

Answer for DFI in a separate thread in RaceTalk