Horses are rated in a handicap to determine their ability in weight of pounds and therefore giving each horse an equal chance in a specific contest.
This also acts as the system to prove who is the best horse in a generation. Trying to determine this in different eras is a difficult task and comparing the ability of individual horses a near half century apart may seem an exercise in futility. However, when those two horse’s names are Arkle and Kauto Star, intrigue arises.

Timeform have rated Arkle on 212 and Kauto Star on 191 but I have questioned this figure as I can’t believe that a horse could be so far ahead of the greatest horse that I have ever seen.

Having read books and watched replays of Arkle, I can’t deny what a legend and brilliant race horse he was. In the 1965 Gallagher Gold Cup, he carried 12 stone 7 pounds to beat the other top horse of the time, Mill House, by twenty four lengths while giving him 16 pounds on top of it.
In doing this, he also set a track record which still stands today.

Over the passed few years, I have worked on a handicapping system which has now proven to be effective, not only in assessing horses but also in predicting race results. In the 2017 Cheltenham festival, either the highest rated horse or best treated horse through my method won 13 of the 28 races, with seven finishing second, another two coming third and one more in fourth. There were also two races I didn’t preview (Bumper and Foxhunter’s Chase) which meant that only four races failed to see my selection make the first four - Champion Hurdle, Champion Chase, Fred Winter and Martin Pipe. The horses were Labaik, Altior, Singlefarmpayment, Vroum Vroum Mag, Tiger Roll. Gold Present, Neon Wolfe, Might Bite, Supasundae, Cause of Causes/Bless the Wings (first and second), Yorkhill, Presenting Percy, Un de Sceaux, Nichols Canyon, Baron Alco, Let’s Dance, Pendra, Defi de Seuil, Arctic Fire, Wholestone, Native River and Dandridge.

The reason for mentioning these is to give example to my handicap and to prove that it is worth something when comparing Arkle and Kauto Star. I begin by focusing on race speed efficiency which is the difference between average race speed and finishing speed of a performance. The theory behind this is that in order to finish the race strongly, the horse would have had to have slowed down at some point and the finishing speed would therefore match the average result to 100%. This would prove a genuinely run race.
In proving it’s genuineness is not to prove it’s quality and a speed figure needs also to be found in miles per hour. Over the time of perfecting my system, I have found that in general, top class national hunt races on undulating heavy tracks run a consistent 28/29 miles per hour, whereas quick flat tracks can reach 34 miles per hour.

I created a template of these speed figures over a variety of twelve races where the resulting winning figure has been proven to be accurate. This allows me to give a performance figure to a maiden horse and that is a method of which is rarely wrong.
For example, Labaik’s maiden hurdle win in 2016 came out on 159 with me, whereas Melon’s figure for his debut success was 156. In the Supreme Novices’ Hurdle, Labaik beat Melon by three and a quarter lengths and Labaik went on to finish about three lengths behind Wicklow Brave (who I had found to be 162). So, through my foundation system of assessing horse’s, Labaik has been consistently 159 with Melon equally beaten as far in the Supreme Novice Hurdle, as I predicted. It’s fascinating that this figure was found in maiden hurdle wins.
By employing this method to Arkle’s 1965 Gallaher Gold Cup success can show what was achieved by the great horse. First, we need to find what level did Mill House run to on the day. I keep hearing and reading that Mill House was reported to be back to his best that day, as if to suggest that he was but it can’t be the case. Arkle beat him by 24 lengths while giving him 16 pounds so in handicapping terms this is forty pounds superiority. I timed the race at 5 minutes 19.8 seconds from the starting tape to the winning line and also lapped the race from the three furlong pole, where Arkle and Mill House are upsides around the Pond fence. From this point to the winning line it takes the winner 45.51 seconds (which averages at 29.66 miles per hour). The same distance takes Mill House 50.79 seconds to cover (26.58 miles per hour). Arkle sets the track record by covering the three miles at an average of 33.27 miles per hour which means that the race record is actually only 87.69% efficient. This basically means that there is an inconsistent speed of the winner, throughout the race.
It would seem ludicrous to suggest that a horse can have an average race speed of this quality while carrying 12 stone 7 pounds on his back in a record time but there is more to the Gallaher Gold Cup when reviewed more closely that explains the fast time. For the first four fences, The Rip and Candy sprint at the front of the field before the front running Mill House takes over for the railway fences and up to the Pond fence for the first time.

From here, the free running Arkle takes it up until they reach the railway fences again, where Mill House goes into as much as a four length lead. As mentioned, Arkle picks it up again at the Pond fence before kicking clear of his rivals. The numerous changes of horses making the running allow for a constant high speed but also allow for each other to have time to fill their lungs as another horse picks up the running, particularly when the shrewd Pat Taaffe allows Mill House to kick four lengths clear down the back, second time around. Horses in a race will travel at the same speed but the ones to the front will also pull harder as they race away from the other runners, so although the horses getting a lead are running as fast, they’re not burning up as much energy as the leaders. The unorthodox running at the front of the Gallaher Gold Cup field explains the track record.

The finishing speed of Arkle is also very good but it’s not an efficient speed figure - meaning he has slowed down significantly at some point, possibly over the railway fences for the second time. It still means we have to compare Arkle’s speed figure to modern day handicapping.

The question now is how good was the opposition on the day in terms of the race performance. By using the method to find a maiden win, we can have some idea of what they did in the race. I do this by taking an example from the Punchestown Gold Cup 2017, where on a similar style of track, Sizing John performs to a 169 figure by sustaining a 100% speed efficiency of 28 miles per hour. No matter how efficient, fast or how much weight Arkle had, it is obvious it was a sensational run. Therefore, it’s through the beaten horses which we judge the race. In the case of Mill House, his speed comparisons of 33.27mph average race speed to 26.58mph finishing race speed is a 79.82% efficiency.

HIs average speed is 114.72% of Sizing John’s which makes him 193.87 rated but only 79.89% of that is accurate racing, which makes his performance in the race 154.88. W e also need to know the correction of Mill House’ speed at the weight - which is 11 stone 5 pounds. Dividing the template weight in the example, of 164 pounds (11 stone 10 pounds) into the rating and multiplying 159 pounds (11 stone 5 pounds) gives a re speed for weight rating of 150.15 for Mill House.

Arkle beats that performance by twenty four lengths with sixteen pounds more on his back which makes him 190.15. This is 22 pounds below Timeform’s opinion but nonetheless a huge rating.

When Kauto Star beat Madison du Berlais in the 2009 King George VI Chase at Kempton, he redefined how race distances were published. Previously, an almost unheard of 30+ winning distance was just regarded as a ‘distance’ but after this race it was accurately stated. It was 36 lengths between first and second with Kauto running a 99.62% solid race. Madison du Berlais’ race time is 6:05.23 which is an average 28.94 miles per hour against a finishing speed of 28.57, which is 98.72%. The British Horse-racing Authority’s official pre-race rating of the runner up was 169, of which he runs to 98.72% making him a 166.83. Kauto is 36 lengths better than that which puts him on 202.83.
It is clear that the King George VI Chase of 2009 is a more solid race in terms of speed efficiency than the Gallaher Gold Cup of 1965. By judging Madison du Berlais with my template example of Sizing John, as I did to determine Mill House performance figure, gives a 100% identical result to the method of using race speed efficiency percentages of Madison du Berlais’ rating.

This not only backs up my review of the two horses but also shows how accurate my system to review race performances of horses is.

I have a few other points to make about the Gallaher Gold Cup. Firstly, how accurate is the race distance? We are told it is three miles but the speed figures don’t match that in my opinion. It was only last year that Wetherby was found to have inaccurate race distances and a review of the entire system brought significant changes. If it’s not a genuine distance, then Arkle may have recorded a more solid race speed efficiency. However, it would counteract his superiority figure of average race speed.
The change in the front runners in the race was already mentioned and racing style is an example of this. Keen horses are dropped in to get them to settle as are horses who are ridden for the trip, whereas horses with plenty of stamina are ridden from the front as they can race with a grab of the bit but still have the energy to stay on. This shows that the four changes up front at Sandown in 1965 allow Arkle to thrice settle in behind of the stronger speed which may back up the poor efficiency score - quickening and slowing throughout the race.

I note that modern day race times are started when the tape is raised and horses could be fifty yards from the race start at this stage, whereas in the 1965 race, the horses noses are almost on the line when the race goes off.

Combining these might explain the strong track record, or indeed individually they could support it. Also, consider the level of horses competing over three miles at Sandown in the modern era, there is no three mile Grade 1 and the highest class of staying chase at the west London track is the Oaksey Chase (grade 2) over two miles and six and half furlongs on the last day of the season when nearly all of the top horses are either done for the season or are in Ireland to race in the Grade 1 Punchestown Gold Cup. The 2005 King George was run at Sandown, as Kempton was laying down an all weather track. Kicking King’s official race winning time was 51.72 seconds slower than the time I recorded for Arkle’s track record, which translates as a 232.74 pounds (51.72 seconds x 4.5 lengths per second) inferior performance before you add on the eleven pound weight difference. This would make Kicking King run 243 - 244 lengths behind Arkle. Another reason to question the race’s genuineness.

It will never be 100% proof but it is a strong case for Kauto Star over Arkle. The ability to back up the argument is a foundation to a debate, something too many people in racing don’t elaborate on but regardless of who was the better horse, it is fair to say that it will be very difficult to find two race horses who can perform to the sensational level of Arkle and Kauto Star. They say Kauto Star never carried big weights to dish out wide margin wins to the opposition but why should he have to when in the modern game there are so many Grade 1’s to contest rather than risk defeat. Furthermore, his performances off of levels were at times devastating to his rivals - also winning a high class Gold Cup by 13 lengths. Arkle had fewer options, however, but such was his superiority in his generation that they couldn’t give him enough of weight.

Their C.V’s back up their performance figures - Arkle’s three Gold Cups, two Hennessy’s and a King George to Kauto’s five King George’s, two Gold Cups and two Tingle Creeks are testament to the unique talents of these fantastic animals, some half century apart.

Article by Mark Keating