A Yardstick for Boats
Which is a faster raceboat, the Melges 24 or a Quest IACC? What about the Mesh Shop VO-70 Volvo racer; is it faster than the WildWind VOJ-70 version of the same boat? Can either boat beat the new Wildcat45?
For that matter, which boat wins when a gaggle of Trudeau Classics descend on a raceline? Is it the T- One, the New York 30, Epicurus, or the just-released Francois J? Where does the intrepid Flying Tako 3.3 or the Flying Fizz 3.x fit in?
These existential questions try a sailor’s soul; they lead to many sleepless nights and quarrelsome days.
It’s not easy comparing boat performance. There are dozens of factors that contribute to practical boat performance, and sailors in RL and SL spend long evenings debating the relative merits of different features. A boat that is very fast on a beam reach could easily be a wet dog when pointing to wind, and a boat with a spinnaker might well beat a similar boat that’s less equipped to handle a downwind leg. But even the speediest sled won’t win a race if it can’t carve a decent turn.
In real sailing there are several different schemes for rating performance and handicapping a race fleet. A common one in the United States is called PHRF (Performance Handicap Race Fleet).
SL PHRF Hotlaps
Given the need and the tradition to handicap different race boats, back in 2007 a bunch of SL sailors got together to generate a practical, fun, and valid performance rating system.
We called it “PHRF Hotlaps” back then, and the concept was simple. Any sailor could try their luck sailing an easy, 8-12 minute solo test course that included windward, leeward and reach legs plus at least one sharp turn. All the skippers used the same wind parameters to sail hotlaps, and they kept track of their “average, good” lap times scores on each course.
Sailors then posted their lap times online, and those results were transferred to a database. The distribution of individual lap times was evaluated to make sure the samples were representative, and then all the scores from each boat class on a particular course were averaged together. That made it possible to generate simple, statistically valid comparisons of boat performance across the whole fleet.
Here are the first six Handicap Hotlaps courses we used to test boats back in 2007-2008.
In 2009-2010, we continued getting Hotlaps data from many sailors in Blake Sea using the Madaket raceline (see the lower right image). Cynthia Centaur even automated the whole system (woots!).
The lap data turned out valid and reliable. If ten sailors each sailed a Hotlaps course and the average of all their lap times showed that Boat A was 20% faster than Boat B, you could be confident that a different ten sailors on a different Hotlaps course would find the same thing. Boat A would be faster than Boat B on that new course, and in fact Boat A would be 20% faster. In other words, the Hotlaps results had strong predictive value regarding boat performance independent of the sailor or the course.
By the time we finished the Hotlaps project in the summer of 2010, the database contained several thousand lap times submitted by over ten dozen skippers, sailing more than forty different boat types.
The Figure to the right shows the list of boats tested up to January 2010; the list below it adds a few more boats tested between January and April of that year. For each boat, the “Lap” column shows the average of all the posted lap times for the Madaket Hotlaps Course.
To make comparisons easy we then chose one boat, the Trudeau J-Class, as an “Index” and used it to calculate a Handicap Factor [the ratio of (Index Boat Lap Time) / (Test Boat Lap Time) ]. The last column in Red shows that Handicap Factor for each boat in the list.
For example, the Tako 3.3 had a Handicap Factor of 1.32, meaning it was 32% faster than the J-Class on the various test courses. In contrast, the Shelly Fizz had a Handicap of 0.52, meaning it was only 52% as fast as the J-Class standard.
Using the Handicap Factors, it was easy to compare any two boats based on their lap performance. Perhaps more important, Handicaps made it easy for a sailor to compare his/her own lap scores over time, and to see how their own results ranked up alongside other skippers in the fleet.
I started thinking about Hotlaps again when aakagon Resident (aka “Topsail”) contacted me recently. Topsail is a serious sailor in real life, and he thought it was important for SL to have a way to compare the performance of different boats in order to handicap mixed fleet races. He discussed this with MarkTwain White, and then sent me a note arguing in favor of a time-based ranking system that sounded a lot like Handicap Hotlaps.
Topsail was right. The last Handicap summary I posted was in April 2010, and very few of the 40+ boats on that list are still sailing in Second Life now. It made sense to start Hotlaps rolling again. It would at least be a good excuse to have some fun sailing solo laps on different courses, while we all compared results and pooled the data.
In the past two weeks I’ve discussed Hotlaps with Kentrock, Hawk, Elbag and others to get their sage advice, and they’re on-board to help re-launch the project. However, Hotlaps is open to everyone of course, and the results only get better when more sailors and racelines are involved; so please drop me a note if you want to set up your own Hotlaps course, or you want help doing it. Either way, we’ll plug you in to the new spreadsheet.
Speaking of those details, Hay Ah’s offered to help refine the interface to work better with her raceline. That should be very interesting to see.
Over the next day I’ll post specifics about the initial Handicap Hotlaps 2013 race courses and the line locations, as well as the details a skipper needs to run a Hotlap. I’m guessing this could be a lot of fun, and something easily accessible to sailors of all skill levels and time zones. (You got ten minutes? You could be sailing a Hotlap!)
Handicap Hotlaps Link History
Here are past links to articles, discussions, and data threads about SL PHRF Handicap Hotlaps. Since the topic stretches back nearly six years, many of the early links are broken; I’m still trying to recover them. I know there are also many more posts on the topic than what I’ve listed, so please let me know if I missed a thread that you think should be included.
Feb 11 2007 Handicapping
A post that discusses handicap issues for the then-new “Big-Boat Races.”
Apr 5 2007: Handicapping sailors
A discussion of ‘handicapping sailors’ versus handicapping their boats.
Nov 20 2007 Should big-boat races be split up?
Big boat racers decided the newly-released Larinda was significantly faster than the equally-matched Yawl and Defender II. How much faster? J Trudeau, helpful as ever, said she didn’t know and suggested sailors figure it out. I think this thread is the first place the basic ‘rules’ for PHRF Handicap Hotlaps were listed.
Nov 25 2007: Deviant Hotlaps
A discussion of wind variation and its implications for valid, reliable lap time scores. It degenerates into a discussion of whether the sailors are the ‘deviant’ ones…
Nov 26 2007: Trudeau Handicap Hotlaps
Nine pages of laps, data, discussion about performance handicapping.
JANUARY 3 2008: PHRF Hotlaps summary
Feb 3 2008: Adding It All Up
February 2008 update of lap data and handicap scores.
Feb 6 2008 (and many updates): Handicap Hotlaps Results
This is a sticky thread with spreadsheets and summary tables for the different test courses and boats in 2008. (The spreadsheet shapshots are currently offline)
FEBRUARY 11 2008: PHRF Summary
Jun-Oct 2008 PHRF Handicap Hotlaps
12 pages of discussion and data from 2008
AUG 7 2008: A New Handicap Hotlaps Course?
A short thread discussing issues with the past phrf test courses, and new ones are proposed and tried at NYC and FIYC.
AUG 10 2008: FIYC Hotlaps
PHRF discussion and data on Epicurus Emmons’ FIYC course.
The results paralleled all the others.
Mar 2009: PHRF Hotlaps 2009
Following the move to Blake Sea, Cynthia Centaur, Francois Jacques, and Jane Fossett set up the Madaket Hotlaps Course with a user-friendly web database system.
Apr 2009: Madaket wind setter out of date – why?
Apr 14 2009: PHRF 2009 Madaket Hotlaps Discussion Thread
Discussion of the new Madaket web-based hotlaps system.
June 2009: June PHRF Update
Summary of the Madaket data compared to 2008.
July 19 2009: PHRF Update, WildWind
More Madaket data, discussion of WildWind boats, concern different wind engines could yield less reliable comparisons.
May-July 2009: RCJ-44 and JMO-60
Specific discussions of performance data for these boats:
August 11, 2009: Discussion of Wildwind apparent wind factors and impact on testing.
November 20, 2009: Summary results for different boats.
April 12 2010: Summary results for 2009 http://metaversesailing.net/2010/04/12/phrf-update-2009/
April 20 2010: J-Classic used as new index boat. The fleet results were still consistent, and many more boats and data points were added.