I am not a number
by Dougal Henshall 14 Nov 2017 12:00 UTC
14 November 2017
I am not a number © The Prisoner
You probably have to be at the very least middle-aged to fully appreciate the tag line reference to 'The Prisoner' TV series, but as that means you're in the demographic majority it is okay to continue.
However, with many a true word said in jest, it does sometimes feel as if the UK domestic dinghy scene is indeed held a prisoner, shackled somehow to the "slings and arrows of outrageous fortune" that is the current system for handicap racing. Given that the RYA are currently championing the cause of increased participation in our sport, it might seem a bit bizarre that some of their recent initiatives have focused more on the ideal of rebuilding single class fleet racing than on improving the lot of the handicap fleet sailor. The operative word here is ideal, for though the concept of one-design sailing has much to recommend it, at the same time the idea of trying to shoehorn an increasingly diverse population into a 'one size suits all' solution seems to smack somewhat of desperation. For it is a harsh truth that in the UK at least, the single fleet racing horse has long bolted, leaving the majority of our sailing at the grassroots level to take place within the handicap environment.
Elsewhere, in many of the other premier league sailing nations, there is far less of a reliance on the use of handicapping systems, though it should be noted that many of these sailing authorities are now becoming increasingly aware of the fragmenting impact of the so-called 'new' classes.
However, rather than denigrate the use of handicapping as it is currently applied, we maybe should instead be applauding just how well it does work. Indeed, it is difficult to see how our sport would have developed in the way it has, without the benefit of a platform that would allow a whole range of different boats to all sail together. And, to be fair, for much of the 65 or so years of the 'Portsmouth Yardstick' system, handicap racing can be seen to have worked very well.
To understand not just how the system works, but to grasp some of the nuances of handicapping, requires us to go back to the early 1900s and the arrival of dinghies designed solely for the single purpose of racing (note to the purists; I am not ignoring racing canoes but they have a different evolutionary path that can be addressed elsewhere). Here in the UK, things really kicked off with the BRA-1 (Boat Racing Association) which would ultimately become the International 12, then the various 14ft dinghies that would come together to become the International 14.
But the problems of diversity would very quickly raise their heads, for down in the far South West, they wanted something a little more robust, which would result in the National Redwing. The early National 12s, in the guise of the Uffa King dinghies, plus the National 18 were just smaller or larger versions of a clearly identifiable family of designs. However, even before development came to a halt in 1939, the dinghy scene was being influenced by developments from abroad. It is easy to think of the hard chine dinghy in terms of Jack Holt, but the Snipe from the USA and the Sharpie from Europe were already setting the scene for the explosion in performance dinghy development that would start in the years following the end of hostilities in 1945.
As early as 1948, there were rumblings of discontent that the arrival on the scene of the new Merlin class (even pre-amalgamation with the similar Rocket) would dilute the existing class structures for the 14s and 12s, despite the former being seen already as too expensive and the latter too small! Yet in truth, the dinghy scene was already heavily fragmented, with local boatbuilders around our coasts building their own versions of what might have already become established in that locality.
A good example of this would be the area around Portsmouth, where the three natural harbours of Portsmouth, Langstone and Chichester had between them spawned in excess of 20 local classes; if you added in the rest of the Solent catchment area, that number nearly doubled. Little wonder with so many small groups of not very disparate boats, there was a demand for a solution that would allow them to all sail together. One solution that was employed in these early days was a time-on-distance handicap scheme, based on having a known 'scratch' boat to work from.
Meanwhile, at Langstone Sailing Club, Stanley Milledge, who went by the splendid nickname of 'Sinbad' was, by 1947, working on a mathematical model using a 'time on time' structure for handicapping. His baseline was the Island One Design, which he assigned an initial position of 100 (in today's money, that's RS Feva territory) with the actual finishing time being corrected to reflect the final finish position. For a while there was a passioned and often vocal debate about the relative merits of time-on-distance versus time-on-time, but Milledge was by now refining and developing his system which was already gaining traction elsewhere. Although under ideal circumstances the time-on-distance platform could be seen as more accurate, it quickly failed to handle a situation where dinghies started to plane, a problem that would become ever more entrenched later in the handicapping narrative.
In these pre-computer days Stanley Milledge must have been an obsessive worker, for he ploughed his way through masses of collected data on dinghy performance and by 1951 had created the Portsmouth Standard Handicap Scheme, and the following year published the first edition of the Langstone Handicapping Tables. These were simple 'look up' tables, covering a range of PYs from 65 to 150, for races ranging from 1 to 3 hours in duration. At the heart of these tables was the simple equation whereby the corrected time for any dinghy in a race would be its elapsed time, expressed in minutes, divided by the Handicap Number, then the result multiplied by 100. Yet despite this being the analogue world, where the finish of a boat had to be caught by watching the second hand on a clock, the system worked well; indeed too well for the vast number of small local classes.
The ability to interleave the multitude of new dinghy classes that were appearing on the scene from the likes of Jack Holt and Ian Proctor into the existing local club structure, would very quickly bring about the first major extinction event in the dinghy world, with this being almost as total a culling as that which saw the end of the dinosaurs. A large number of classes disappeared without trace, though they were joined in obscurity by many of the new designs that were supposed to replace them.
In an amazing period of dinghy development, some new boats scarcely got past the prototype stage before being superseded by something better. The advantage of the Portsmouth Handicap system was that new boats could be launched and sailed with the rest, with this giving an early indication as to the capabilities of the design. Those that had what it takes would succeed and could grow, gaining the critical mass to become standalone classes, at which point would they could be promoted from Secondary Handicap status up to the exalted level of the Primary Handicap list.
Even though the system was working well, it was coming under pressure from a number of developments. Technical changes, such as the introduction of terylene sails (which saw the National 12 PY drop from 86 to 81 on the 1955 scale) needed to be fed into the numbers, whilst the arrival of much faster boats, mainly the one-design catamarans, such as the Shearwater, would question the practicality of racing under the current handicap range.
By the time the country advanced into the swinging sixties, the RYA were running the scheme but, with the reliance on clubs supplying paper-based annual performance returns, numbers tended to be static. The PY system itself was still run on a 'non-intervention' basis, which effectively agreed that every dog could have its day. Two popular Jack Holt designs of the day, the Enterprise and Hornet, showed this aspect of the handicap system off to perfection. Get a day of light and shifty breezes, on flat water and the Enterprise was in its element, easily able to beat much supposedly faster boats; get a breezy day with a sea running and the Hornet was hard to beat on handicap.
Over the course of a season, the best sailors still won, though everybody would remember (and recall when in the bar) the day when it all went wrong... the day that started breezy, only for the wind to die, or the reverse change in conditions, which could make an apparent mockery of the best efforts of the handicap system.
Given that we can now look back at the performance of the early handicap system with the benefit of 30 or more years of hindsight, one clear observation was that handicap racing worked at its best when the range that it had to work across was not too great. Racing Albacores against Enterprises, Solos against OKs gave a reasonably good indication of who was sailing well and who was getting it wrong in a big way. For those who were interested enough to examine the performance of the handicap system, it was clear even back then that the greater the range of boat PYs in any one given race, the greater was the potential for 'distortion' in the results.
Even when problems with PY racing were escalated upwards, the core of the defence was that from the outset, the RYA had stressed that their numbers were nothing more than a starting point. Their declared intention was that clubs should take on a more interventionist policy of applying local corrections to boats that performed consistently above their PY (or consistently below this PY), though the reality was that few Sailing Secretaries risked the anger that would be directed at them when numbers were tweaked.
Other clubs tried to overlay the PY system with the addition of a personal handicap correction, where sailors who invested in their boats or worse still, dared to practice and attend training sessions, would be penalized for winning races. Meanwhile, the vagaries of the handicap system would give rise to the near-myth of the handicap bandit, a boat that was so generously rated that you only had to cross the start line to be assured of a win! Like so many myths, this one had an element of truth about it, with opportunities to manipulate the system being identified – and sometimes exploited.
Many of these involved boats that now resided on the RYA 'Moribund' list of defunct classes; the idea being to take one of these, fit it out with a modern rig, fittings and foils and the result would, if raced on the old handicap number, be a very competitive package. Yachts & Yachting journalists Jack Knights and Bob Fisher were threatening to do just this with a Pegasus (think of a flattened-out Albacore, with more sail area, spinnaker and trapeze and with a modern PY of 1068!) when Jack's untimely death ended the project. Still, there have been those who have played this game to some effect, though increasingly clubs have become wiser of the all too obvious 'bandit boat'.
The other problem that the system had to deal with was the arrival of new classes onto the scene and how they could be given a handicap before there was any empirical data on performance. Throughout the lifespan of the PY system there have been examples of designers and builders suggesting very generous numbers in the hope that a few big race wins would give an added impetus to early sales. Sadly, the opposite was more often the case, when vested interests could combine to see that a new design that might challenge the current niche holders would be given a swingeing harsh PY at the outset. This certainly worked, with some very good, innovative and deserving designs struggling to gain acceptance, soon to vanish; after all, who wanted to buy a boat that was considered a 'dog' (i.e. the opposite to the bandit) when racing on PY.
Even with all of these negatives, for the next 40 or so years, the core PY system, administered by the RYA, was seen to be working well enough. If you didn't like racing on handicap, the option was always there to go away and race single class fleet events, except for the worrying trend that one design racing appeared to be in retreat, as new trends, such as the rise of the single-hander started to dramatically change the landscape of the sailing scene. And it would be one of the oldest singlehanded dinghies that would provide a new, stern test of the ability of the handicap system to respond to change. The International Moth of the 1950s and 60s was nothing more than a smaller version of a normally shaped dinghy, with a relative rating that positioned it slower than a Firefly.
In just a few years, the Moth would change, becoming ever skinnier, more extreme and - in the right hands - incredibly quick. At a windy winter handicap event in the early 1980s, for invited top helms to represent their class, watchers were treated to the spectacle of a Moth sailing past a hard charging 505 upwind (only to self-destruct when bearing away); clearly the old PY numbers were no longer applicable, though in changing them year on year, this ensured the obsolescence of Moth designs that were but a year or so old. The story of the rise and rise (in speed, numbers, and vertically out of the water) of the International Moth is well documented, yet this foiling success shows just how hard it has been for the handicap system to cope with change; the Moth PY has seen a greater percentage change than any other dinghy, yet the class is still included under the umbrella of the existing scheme.
Luckily, help was at hand as first the hand-held Sinclair calculators, then PCs and laptops would come to the aid of those running handicap races – and those who ran the whole handicap system. As sailing clubs belatedly joined the rest of the world in becoming 'connected' the amount of data available on handicap performance started to dramatically change the way in which the handicap system was administered. This resulted in some strange anomalies, with some boats apparently getting quicker in PY terms, whilst others appeared to get slower. The Laser, a boat that - apart from the introduction of the XD controls - should have been a known, fixed quantity, found it's PY number sliding towards the more generous end of the spectrum, whilst other single handers were getting faster year on year.
Of course, if you sailed a new FRP boat with a carbon raking rig then maybe you could sail to a sub 1000 PY handicap; sadly, for those in these classes sailing anything other than the latest state of the art boats, success in PY races was jacked up out of reach unless their club was enlightened enough to introduce a sliding scale of PYs. These changes in the handicaps were the output from the RYA's PYAG (the Portsmouth Yardstick Advisory Group): the body who each season are tasked to examine the collected data and issue revised PYs on the morning of the Dinghy Show.
It would have been nice at this point to be able to say how this process of interpretation of the previous year's data is conducted and what factors are being considered when the existing PY for a class gets changed. As part of the research into this article, the RYA was one of the bodies that were polled for information, along with not one but two polite, formal requests to be allowed to attend the PYAG sessions as an observer. Sadly, both these requests met with direct refusals, so the output of the PYAG must be taken for what it is.
Luckily, there are alternatives that can be examined, where transparency allows us to see how the availability of historical data can lead to a more focused policy on setting PYs for various classes. The SailJuice series of UK-wide winter events is now one of the main powerhouses of the UK domestic dinghy scene and they are very upfront by clearly displaying what Great Lakes handicaps will be in use for that year.
Moreover, the Great Lakes Committee maintain a website (greatlakes.org.uk/The-Method.php) in which they clearly document how the numbers in use for the current season are arrived at. By stressing how they set handicaps based on the potential for the boat, excluding the scores from poor performers, the SailJuice Great Lakes numbers are probably the closest yet to a real measure of performance for this particular set of conditions. Analysis of the results from previous seasons racing suggests that they are close to achieving their aim of ensuring that those who win do so solely as a result of skill, not as a result of any manipulation of the boat's handicap.
As is clear in the naming, the Great Lakes numbers have been arrived at working from data arrived at from races held on large expanses of inland water. Meanwhile, Hayling Island Sailing Club have shown how PYs need a more intuitive approach at locations where the tidal effect is a major consideration. The arithmetic on this is pretty straightforward; it is inescapable that a current flow has a significant effect on handicap performance. The faster the boat, the less the impact tide has on performance on handicap; it isn't just theory but a practical reality that, if facing a beat into a strong tide, a 49er would take longer to reach the mark whilst a Mirror might not make it at all.
The Great Lakes and Hayling Island might be just two of the innovative approaches being taken to the issues surrounding racing on handicaps, but both these are still fairly modest amendments to a core system that is now well past pensionable age. This then begs the question, "is this really the best that we can do"? In a sport that has been so consistently rich in innovation, is it really beyond the imagination of the combined intelligence of either a clever individual or a group of smart thinkers to come up with better ways of sailing disparate designs of boats together on a single race course. Simply recognizing that the PY works better within a narrower band would be a start, for no matter what the mechanism used, racing a foiling A Class cat against an RS Tera is bound to be problematic.
Were there to be a search for a better system, in all likelihood, it would be along the lines of a time-on-time basis as this has been shown that it can work, but one improvement could be in the way that the base PY numbers are originally arrived at.
Forward thinkers such as SailRacer have already collected large amounts of real time on water data, thanks to the tracking units that they can attach to boats. Taking their enlightened approach a stage further could see an element of the original 'time-on-distance' handicapping, where across a prescribed range of conditions and sailing angle, the true boat speed is assessed in both planing and non-planing modes. It would require a very disciplined and exacting process of data collection, but when used to further refine numbers such as those in the Great Lakes listings, could well see a more believable number that could then remain static unless there was a modification to the boat that impacted on performance (such as the new top section and sail for the Laser). One of the advantages of this methodology would be that when such modifications were made, or new designs of boat launched, being put through this process would produce a PY that carried with it a better degree of reality than the guesstimates of some Experimental numbers.
However, whatever the system used and despite the title of this article, we really are still tied to a number. It matters not if, when signing on for a race, you write down Fireball or RS400, somewhere along the line, that information is turned into a number. Most entries to races under the handicap system even still demand that the PY number be supplied; Name of Helm (and crew if a double hander), Class of Boat, and then - PY. Practically, we don't sign on as a class or even as an individual, we sign on....as a number! But what if this were not the case?
As part of the research into PY systems for this article, I also sought out some of the software developers who have more than a passing interest in how the use of IT at sailing clubs could radically change the nature of handicap sailing. As it was made clear to me, the use of a PC-based results system such as Sailwave hardly causes the processor to break into a sweat, such is the power resident in most modern computers. Even with a budget level machine, there is more than enough capability inside the box to drive a far more detailed handicapping platform.
Many sailing clubs now have their own weather station and even those that do not ought to be able to access a range of nearby sources that accurately record wind strength and direction. If the club is sea based, there are programmes now commercially available that predict not only the tide times for years to come, but direction and strength of flow. Given a normal club race duration of something between 45 minutes and an hour, if these sources were polled at the start and every 10 minutes after that, the 'system' would have a very detailed picture of the prevailing conditions the competitors would have experienced when racing.
The next element would come with the layout of the course, with many clubs already drawing down from a library menu of courses held on the club PC; enter wind direction and strength and the PC prints off the course size and orientation. Associate this information with the wind data that is being read and the computer will know how much of the race would be spent beating, running, close reaching and fetching and broad reaching, up, down and across the tide. All the pieces are now in place for the 'system' to apply some not too difficult rules to come up with a meaningful PY for that boat, on that course, racing in those conditions.
Better still, the software would include the ability to be 'tweaked', so that if you raced at a location where there was an overriding set of conditions (say, on a river where the beat would entail a great deal of short tacking) the system would recognize this and alter the PY accordingly so that a fast tacking boat such as an Enterprise would be impacted to a greater degree than a Hornet. The system wouldn't have carte blanche to determine PYs for any given class, but instead would select the most appropriate from a pre-determined range, based around a core number +/- 'x' points. It all sounds rather techy, but in truth it isn't, because we all know what conditions our particular boat sails best in. Today we determine our PY number before we go afloat, not knowing what the conditions MIGHT be at the finish. Is it really so radical to sign on as a Fireball or a Feva, a Solo or a Scorpion and have your handicap assigned to you on the basis of the conditions you actually sailed in, rather than those you think you might get (although one group who may not like it might be those who can pick and choose their rigs, depending on the conditions of the day!).
Importantly, there is nothing in this possible solution that is space age or awaiting the next big thing in technology, for everything here, in one form or another, is available now. The only downside would be in the entrenched resistance to loosing that connection between your class and it's PY number – for that year. Yet this was just one 'off the wall' idea, suggested by an individual for whom the challenge was in making something like this work, rather than what it would mean to his success or otherwise out afloat – there must be many more ideas out there, still waiting to be heard. But if those who run sailing truly believe that there is nothing better than what we have now, then the door will never be open to these new ideas.
But in a sport such as ours, innovative ideas like this aren't just interesting, they are now essential! When I made my request to the RYA for the chance to attend the PYAG, at the same time I asked if they were actively looking at alternative methodologies aimed at enhancing the PY system. I have no doubt that you'll be surprised that this too got a negative response, which has to be a major source of concern.
For like it or not, at the grassroots level, we're stuck with handicap racing. The initiatives that have come down from above have been more aimed at penalizing the handicap sailing, whilst rewarding the fleet sailor; who can forget the suggestions that clubs should give the best boat park spaces and only allow socials for those who agree to sail in a one design fleet? Yet the name of the game now is 'participation', the in-term for now that translates into getting more people out afloat.
Maybe relying on that suggested stick, rather than the 'doing what we do now but better' carrot hasn't shown to bring about a dramatic increase any more than the loved at the top concept, that the major stadium events (Olympics and Americas Cup) will bring in a whole new generation of sailors.
There is of course that current response in which we do nothing, preferring instead to stick with the same old flawed and often discredited PY system. But as we will see in Part two of this three-part series, the Niagara Falls-sized cliff edge, that the sport is drifting towards on a one way tide, is a very real threat to its continuation in its current form.
This first article started with a quote from a famous TV series, so I will let you guess what the tag line for Part Two will be. "We're all........."