All boats go at speeds dependent on such things as engine or sail power or oars, boat weight, crew weight, length, width and hull shape. If the boat is designed to travel flat with even shape in the water either side of the centreline, then travelling heeled over will be slower. This latter point is the case with all centreboard dinghies so once you are leaning fully out, it is faster to let the sails out (reduce heeling ) even if the sails flap a bit. With keel boats which are big enough to live in or go round the world in some cases, the amount of sail power is more than the available righting power of the lead or metal keel under the boat and they have to be designed to sail to windward heeled over. Even then you will see photos of a row of crew with their legs over the side sitting up to windward like a dinghy .
In light centreboard dinghies like we sail, the hull shape and weight produce big differences of speed and so does the sail area , such as using a spinnaker or getting more leverage for the crew using a trapeze thus allowing more sail to be carried. The Albacore for instance belongs to the post-war era of traditional looking hulls but without a steel centreboard to help keep it upright, thus reducing weight and allowing planing. When all small boats looked like that it was fairly easy to estimate the different speeds that 14ft, 15ft, and 18ft boats would go at and after some trials handicap figures were arrived at to enable mixed fleets to race in say Falmouth or Portsmouth Harbour. In those days, cars were scarce and boat trailers non-existent so travelling to anywhere except by water to the next river club was unknown.
As time went on and more money came into the sport, designers tried adding spinnakers, trapezes, very wide boats to help sitting out, very flat hulls to aid speed off the wind and soon there was an explosion in designs all with slightly different speeds even when they had the same length and sail area. So handicapping became a complex performance but not much more scientific. In fact by now it consists of gathering results and best guesses from all over the UK on all sorts of waters and number-crunching to produce some sort of average - see how it goes and then change it a bit next year if the results so indicate.
The original owner of the Portsmouth Yardstick copyright (because in the beginning it was invented to allow handicap racing in that harbour) who rejoiced in the name of S. Zillwood Milledge, sold Handicap tables rather like the old school logarithm tables. This was in the days before calculators so you could do multiplication without too much brain overload, but long division was something again and to do the calculations for twenty boats without the tables would take hours. I may be misjudging the copyright holder but he decided to use handicap figures for any class of boat which have to be DIVIDED into the boat's elapsed time for the course. If he had used handicaps which could be MULTIPLIED by the elapsed time, it would ( and could to this day ) have been easier but he might not have sold so many tables.
By the present day, we know that the type of course makes a big difference to a particular class' speed especially catamarans (which were not thought of when the PY scheme started) and any craft which goes much faster off the wind than upwind. To a great extent, beating speed is still dependent on waterline length unless you add enough sail and righting power (double trapezes, say ) to plane to windward and even then you won't do it in Force 1-2. So if we all did beating races and then finished, results would be more predictable. But as the older designs turn the windward mark and this includes unassisted single-handers, the change in speed is quite modest or even slower than beating until higher wind strengths when capsizing is part of the equation. The newer faster designs do this jump in speed as they turn off the wind at much lower wind strengths and you can see this effect in the results for Force 1 days as opposed to Force 3 plus days.
Because of the increasing speed and complexity of classes, the RYA (now owners of the PY scheme) recently changed to four figure handicap numbers which gives a quite undeserved impression of precision over the old three figure system. From what I say above, you can see that you are lucky to allot a number to within 2% which is 20 points in a 1000 and yet we have intense debates over 3 or 4 points in a 1000. Clubs are encouraged to use variations of the National numbers where experience over a period of time shows that a class (not particular helms) is generally slower/faster than the National number. We have done this mainly for the smaller boats who can suffer in light airs when surrounded by many sails in our relatively restricted water, but we have found it is only safe to do this adjustment when a class is turning out a good number of boats which can be truly said to be average club sailors - not "stars" or "novices ".
Now for some arithmetic. If an Iso takes 60 minutes to finish a race, we first turn that time - called his "elapsed time" - into seconds, which is 3600 in this case. The Iso Portsmouth Number (PN) is 926 and we use this to find his "corrected time" by dividing 3600 by 926 and multiplying by 1000 .
So CORRECTED TIME = ELAPSED TIME ¸ PN × 1000
In our example, corrected time = 3600 ¸ 926× 1000 = 3,888.
If in the same race a Laser ( PN = 1078 ) had an elapsed time of 69 minutes 51 seconds then his elapsed time would be 69 × 60 = 4,140 +51 = 4191 seconds and his corrected time = 4191¸ 1078 × 1000 = 3,888 which is the same as the Iso producing a tie. This gives some idea of the relative speeds of the two boats in theory and shows that a Laser must be no more than 9 minutes 50 seconds behind an Iso to beat it if the Iso finishes in an hour. Keen readers can try working out how far a Solo (PN = 1166) and a Topper (PN = 1315) can be behind an Iso to beat it.