After the 2018 regatta with 22 boats from seven countries it was clear that piggybacking on the Swedish SRS rule and extrapolating across it and across ORC to establish handicaps had reached the end of its usefulness. The process had become overly complicated, time consuming and unwieldy. In addition, many of the new participants from Italy, France and Austria sailed with other boats than those common in the Nordic countries. Only few of those were rated by SRS.

     The Göran Schildt regatta had to come up with something simpler and less demanding on the small GSR volunteer organisation.

     The major rating systems (IRC, ORC, SRS and others) keep their formulas secret but there are a number of open source handicap rules. There are the German KLR, the Norwegian KTR, the Mediterranean C.I.M., and others in Denmark and England. Sweden has since more than 10 years an open source handicap rule called the Alma rule used mainly for older and classic boats. It uses a formula similar to that of KLR, KTR and others but has also other features of particular interest for the Göran Schildt regatta:

• it can rate many different sizes and types of boats,
• it needs a limited number of parameters which require none or very little measuring,
• it uses a direct rating process through a website,
• it calculates a rating with minimum staff involvement,
• it delivers a rating in the form of an easy-to-use time correction factor, and
• it is free of charge.

     Like the above mentioned and many other open source rules it uses a formula based on waterline length multiplied with three ratios – namely sail area to displacement, length to beam and draft to length. The relevance of these ratios for determining boat speed potential is universally accepted by yacht designers and has been extensively proven and documented through science and research. See for instance Principles of Yacht Design by Lars Hansson and Rolf Magnusson. The Alma rule and how it was developed is described in both Swedish and English on the Almatal website.

     The rule used for the first time in this year’s Göran Schildt regatta is a simplified version of the Alma rule modified to fit the particulars of modern cruising boats. Features for rig complexities of classic boats have been eliminated as well as age related rating discounts. But our version of the rule retains keel configuration features with added profiles for bulb keels and allowances for folding propellers and bow-thrusters. The Alma rule's aspect ratio calculations have also been kept, however, in a simplified form. The Alma rule has an overly simple spinnaker factor. It has been replaced by a sail area correction factor that also takes into account genakers and code zeros.

     The values of the modified Alma rule’s sail and displacement correction factors were calibrated against a sample of 34 ORC rated boats ranging in size from 51 to 28 feet. Data for these 34 boats, taken from their ORC certificates, were fed into the revised Alma rule formula. The resulting ratings were compared with the 34 boats’ respective 2019 or earlier ORC ratings by way of correlation. The high initial correlation – over 0,95 – suggested that the two rules are based on the same basic formula. A sample of 34 SRS rated boats was also tested in the same manner. Its correlation with the revised Alma rule was lower about 90% – but still significant. SRS is obviously also based on the same basic formula.

     An optimization feature of the Excel program was used to calibrate each correction factor step by step up to a point where the correlation no longer improved. When all factors had been calibrated the correlation between the two sets of ratings had reached 0,97996. With this testing and calibration process the rule for the Göran Schildt regatta had acquired features and a purpose that set it apart from the Alma rule. It had become a GSR rule with a purpose of providing ratings in close similarity to those of ORC – the world’s largest rating system – without the latter’s costs and complexities.

GSR versus ORC for two regatta boats

     GSR versus ORC can be tested for two of the boats in the 2021 GSR fleet – Robust and Trinity – as both have ORC certificates. The ORC time correction factors are 0,845 for Robust and 0,935 for Trinity. The corresponding GSR ratings were 0,910 and 0,991 respectively.

     Table 1 below shows the result when elapsed time in minutes, for both boats and the three races, has been multiplied by respectively the two boats’ GSR and ORC time correction factors. The outcomes are reasonably close – that is Trinity having 102% longer corrected time in the first race in case of GSR and 104% with ORC, 110% to 113% longer in the second and 102% with GSR versus 105% for ORC in the third. This suggests that the GSR rating formula, for the conditions that prevailed during the regatta, produced a more conservative results than ORC would have yielded but that both rating rules produced the same outcome.

Table 1

Note that the rating produces time differences in relation to a balancing point of 1 at a chosen position towards the centre of the fleet. The absolute number of corrected minutes are therefore unimportant.

Updated comparison of GSR versus ORC ratings for 37 boats in reference sample

     The GSR rule used in this year’s regatta had a few bugs which were discovered and corrected prior to start. The corrected formula has thereafter been tested against the reference sample of ORC-rated boats. The sample was increased by three more boats to match participant boats in the regatta. The GSR and ORC ratings calculated based on boat data from the ORC certificates are shown in Table 2. The correlation between the two sets has increased to 0,9856.

     The average difference between the GSR and the ORC ratings is 0,060. Boats for which the difference is lower than the average, such as the two First 44.7, the Bavaria Match 42, the Wasa 42 and the two Bavaria 39 and 38, get a more favourable rating by GSR than by ORC. The smaller boats, on the other hand get a higher rating by GSR. As many of the smaller boats are also older boats this may, in part, be a consequence of the fact that GSR, as opposed to ORC, does not take boat age into consideration.

     This effect is also visible in the graph below which shows the gap between the two curves widening towards the lower left hand side. It should be noted, however, that ORC ratings for the same boat type can vary considerably – much more than in the case of GSR. The table as well as the graph also show the ratings for the two regatta boats with ORC ratings. The First 456, albeit being a relatively old boat has relatively lower rating under GSR than with ORC whereas the opposite is the case for the Rhodes Reliant.

 Table 2 (data for the above graph)

 Pairing and comparing same type of boats

     The sample of ORC reference boats also makes it easy to compare regatta boats with an ORC-rated boat of the same type as is done in Table 3. Ten of the 2021 GSR regatta boats are shown with names in the top row next to an ORC boat of the same type.

     The table shows, in the first and second columns, how the different components of the rating formula are defined and calculated in a step by step process ending in two mathematical arguments (Terms 1 and 2) that are multiplied with each other to yield – in the fourth row – time correction factors for each boat.

     The differences in GSR ratings for most boats reflect to a large extent differences in sail area which is the most important parameter for both rating rules. In some cases the difference appears first with the corrected sail area (shown in the SAc row). The correlation for sail area and rating is 0,89 in case ORC and 0,91 in case of GSR. For both rules correlation increases as sail area is multiplied by the correction factors.

     For GSR to yield fair ratings it is therefore important that sail areas are correctly reported. The GSR program contains automatic checks on those and other parameters. As a result two boats could lower their ratings as their sail area was found to be incorrect.

     The second most important parameter for both rules is the waterline length which correlates to 82 and to 81 percent with respectively ORC and GSR ratings. The GSR rule uses the design waterline length whereas ORC uses what is called IMSL (International measured sailing line) being an average of three different waterline measures. IMSL differs slightly from the design length.

     The largest difference is in the case of the two Bavaria 46 of which one has a higher masthead rig but significantly smaller sail area as compared to a lower fractional rig with a 30 percent larger total sail area. After correction for aspect ratio the sail area difference becomes 20 percent and a rating difference of 6 percent

     For the two Wasa 410 a sail area difference of 10 percent is the main cause of a rating difference of 2 percent and for the Wasa 42 boats the rating difference of 4 percent corresponds to a sail area difference of 16 percent.

     In the case of the two Odyssey 479 a larger sail area for one has not resulted in a lower rating. The combined effect of a small difference in aspect ratio – reducing the sail area difference by 50 percent – and fixed propeller and bowthruster reducing the displacement factor have caused the difference in rating. The situation is similar for the Oceanis 473. A likely overstated displacement value more than eliminates the benefits of the larger sail area for Mikimax.

     For the remaining three pair of boats the rating differences are small and due to minor differences in sail area, displacement and rig – this in the case of the two Bavaria 38. The pair of Odyssey 449 have almost identical hull, rig and sail data. It is differences with respect to propeller and bowthruster that explain the rating divergens. For the Elan 450 there are likewise a number of very small discrepancies together with a bowthruster that combines to cause the small rating difference.