Author: Howard Wright
Who said there were three kinds of lies: lies, damned lies and statistics? Mark Twain used the sentiment in his 1906 autobiography and it has stuck ever since, although the fact he attributed the saying to the British prime minister Benjamin Disraeli has been hotly disputed, since it seems not to have appeared in any of the latter’s speeches or writings. So, although the basic theory holds good, even Twain might have been mistaken and made up the reference to suit his argument. That’s the way with statistics. They can sometimes be manipulated in more ways than one, which happened once the International Federation of Horseracing Authorities (IFHA) published its latest set of annual World’s Best Racehorse Rankings a couple of weeks ago.
Since the IFHA released its handicapping panel’s views on a regular basis, at least six times in a calendar year, it was going to take a seismic shift of opinion to disturb the upper echelon of the list that appeared towards the end of November, showing a dead heat for top place between Winx and Cracksman. And so it came to pass. The handicappers’ final deliberations on those horses rated 115 and above, which were thrashed out over the best part of a week during Hong Kong’s international festival at the beginning of December, did indeed fail to separate the Anglo Australian pair on a rating of 130, two points ahead of America’s best, Accelerate. The not unexpected result still provoked the usual discussion among professionals and public alike, driven partly on this occasion by the perceived deficiencies of the pair.
Winx for not having stepped out of Australia in compiling a record run of Group One victories; Cracksman for swerving any ground surface that had a hint of firmness about it. For all that these comparisons were interesting, the major talking point surrounded the extra set of statistics that emerged alongside the World’s Best Racehorse Rankings, which are used to compile a list of the world’s top 100 Group One and Grade One races. Here was real controversy, stoked by the appearance of 31 races in Australia among the 103 that qualified, after a seven way tie for 97th spot. Talk about lies and damned lies. Was racing in Australia in 2018 generally so much better than in Britain, which came next in the list with 19 races, or the US with 14 or Hong Kong with 11? As esteemed fellow columnists Nicholas Godfrey and John Berry proved in Al Adiyat last week, opinion was divided.
‘Possibly’, Godfrey wrote; ‘Definitely not’, Berry countered. Far be it from me to take sides, but a touch of historical perspective might be in order, to explain how a situation has come about that on the surface suggests Australia can lay claim to the best racing in the world. The problem can be explained in a word: sponsorship. Annual ratings were introduced in 1977, when they were known as the International Classifications and covered Britain, France and Ireland. By the time the IFHA took over administration in 2004, Germany, Italy, North America, Japan, the UAE, Hong Kong, Australia and New Zealand had joined the club, and soon South Africa and South America signed up too. The significant change came in 2013, when the IFHA took on a partner in Longines, the Swiss watchmaker, whose products range from the top end, costing tens of thousands of pounds, to the novelty Swatch brand, and whose involvement in international sponsorship has grown to cover all parts of the globe.
Longines loves a presentation ceremony, and if it involves a lavish lunch or gala dinner, all the better. Hence, with the IFHA’s blessing and backing, it began to introduce a number of annual awards, one of which was for the World’s Best Horse Race. The first such award was handed over in 2016, based on a league table derived from the World’s Best Racehorse Rankings and arrived at by taking the average rating of the first four placed horses in Group One and Grade One races, then averaging the results over the previous three years, 2013-15. This is the principle the IFHA uses to judge whether Group races of any kind are meeting the various ratings criteria, on which recommendations for their upgrading or downgrading are made.
It is an eminently sensible system, and on this basis the Prix de l’Arc de Triomphe triumphed in 2016, with a mark of 125.75. However, that was the end of the three year average system. In 2017, when the previous year’s World’s Best Race was decided, just one year’s rankings were used to judge the winner. The Breeders’ Cup Classic came out on top, while a substandard Arc de Triomphe slipped to joint seventh with a ranking of 122. The system of judging on one year alone is now the norm, and the Arc has held sway for the last two years. The change of methodology might have worked for Longines, since it is simple and provides a clean result, but its application as a measure of overall quality in a single jurisdiction or over a period of time is flawed.
One outstanding horse can make a big difference to the fine margins, and this is what has happened to a significant degree with Australia. In 2018, six of the races won by Winx were in the world’s top 58 races, including four in the top 11, all dragged up the table by her brilliance. They were led, perhaps appropriately by the Longines Queen Elizabeth Stakes at Randwick, which reached a level of 123.75, behind only the Arc’s 125. In 2017, when Winx also won the race, it was ranked sixth with a rating of 122. However, in 2016, when she did not run, it was joint 70th, on 117. Add in the upward influence of decent challengers from Europe, including the Godolphin pairing of Benbatl and Best Solution, and it becomes clearer where the World’s Best Racehorse Rankings have benefitted Australia disproportionately in the World’s Best Horse Races’ table.
Had the average over three years been in operation, the Arc would have merited 124.41 for the period 2016-18, compared with 125.75 for 2013-15. By contrast, the Longines Queen Elizabeth Stakes would have been on 120.91 for 2016-18, instead of 123.75 for 2018 alone. On that basis, the Queen Elizabeth would have slipped down the table, and the Dubai Sheema Classic, with an average over three years of 120.92, would have moved up. Even so, Hawkbill’s victory over Poet’s Word, Cloth Of Stars, Rey De Oro and Best Solution made the 2018 Sheema Classic the joint fourth best race of the year with a ranking of 123, enabling it to improve on its 2013-15 average of 120.25. To complete the UAE picture for 2018, the Dubai Turf figured joint 47th on 118.25, and the Dubai World Cup joint 57th on 117.75.
However, the average over the last three years presents a slightly different picture, with the World Cup on 120.75 (compared with 118.92 for 2013-15) and the Dubai Turf on 118.50 (119.25 for 2013-15). Now that’s what I call statistics which don’t lie.