The Commonwealth Games mile races

In July 2026, John Austin reported that the 1500 m races have all been replaced by 1 mile at the Commonwealth Games. This is his follow-up article on the mile races at the Commonwealth Games.

As I noted in Metric Views in July [1], the 1500 m races for men and women were replaced by mile races in the athletics programme of the Glasgow Commonwealth Games (CG), 27 July to 1 August. The original article stimulated some debate, and I was able to respond with my own views at the time. The discussion was more broadly-based than the mile races themselves extending into other events of the athletics programme where imperial units were present. Here I offer further comment which I hope addresses people’s concerns.

Historically, the mile races for men were the first to be run at the CG since 1966. So, ironically, having gone metric for the first time in 1970 (Edinburgh) the CG ceased to be fully metric in 2026 for the first time, also in Scotland! The Glasgow games integrated para-athletics for the first time into the main events, but the organisers were perhaps careful in not having a mile race. Perhaps the reason for this is that in modern sport the athletes and para-athletes are moving so fast that the event has become dangerous with a high chance of collisions as the athletes race to the first bend. This is not helped by the start which in athletics is from a curved startline 9.34 m from the finish line so that the athletes run 4 laps plus that small correction. The men typically run as fast as women 800 m runners, but the latter are in lanes for the first bend, about 140 m. So, there is a good argument for the men’s mile race also to be set up starting in lanes with the 9 m tacked on at the start. Presumably para races could also be run in the same way. For the race itself, the full distance of 1609.34 m was run. Some people questioned whether the races would be over 1600 m instead. Normally, though, when athletics says it’s a mile, then a mile it is. That is despite Seb Coe’s suggestion that the race was easy to understand as it’s 4 laps. Well, it isn’t quite so easy to understand. Now the 1500 m is easy to understand, except for people particularly poor at arithmetic, as that is 3¾ laps.

Regarding the races themselves, I noticed that in the opening round, many athletes (especially the women) had mediocre, by their own high standards, personal or seasonal bests over the mile, indicating that they had run few races over the distance. Qualifying round composition is seeded, and presumably for the CG it was done on 1500 m times, rather than mile times. But perhaps it was not, and this would have complicated the work of the officials and the fairness of the outcome. In the comments to my original article, “m” questioned what symbol would be used. As I speculated at the time, the word “mile” is short anyway and, indeed, that was used throughout the games as far as I noticed.

For the men, the CG record was from 1966 (3:55.34 by Kip Keino, Kenya), and this was improved in the heat (3:53.85 by Cameron Myers, Australia). The final was a tactical affair won by Josh Kerr (Scotland) in 3:54.12 [3]. The women’s race was also tactical, much more so than the men’s with a final time of 4:39.31 [4] but was notable for an Australian clean sweep of the medals. As the mile race had not been run previously, an inaugural record was set in the heats (Claudia Hollingsworth, Australia) of 4:31.17.

The discussion of my original article [1] raised the issue of whether there was a “creeping imperialism” in athletics and raised further questions about athletic performance in general. I don’t see evidence of an increase in the use of imperial units, but we must remain vigilant! At the European Athletics Championships, which took place 10 – 16 August, just a few weeks after the CG, the race walks were over a distance of a half marathon and marathon. There was a marathon run as well. Unfortunately, in the UK, people nearly always refer to a marathon as 26 miles, or even 26 miles 385 yards but that isn’t correct. In the early modern Olympics and in other races as well, the distance was undetermined [5], but it became established at the “classic 26 miles 385 yards” in 1908. This is all a bit of mythology in some ways. The 1908 course was originally meant to be 26 miles but the 385 yards was tacked on to finish on the track opposite the royal box. So, really the distance was “about 26 miles” plus 385 yards which is really “about 26 miles”. Because the marathon race is held on public roads it needs to be surveyed every time for a new course. Even modern techniques can only survey to a metre per kilometre. Nonetheless that classical distance has prevailed but in actual fact has been converted to metric. So the official course distance of the marathon is 42 195 m and the half marathon is 21 097.5 m. Those are not round numbers in either metric or imperial. Some commentators suggested we should move to a round number in km for the marathon distance, such as 42 km or even 40 km, with 21 km or 20 km for the half marathon. This would certainly bring about a rational design in the athletics programme with 10 km on the track and a doubling for the half marathon and another doubling for the marathon. In fact this would fit well with human physiology. It appears that in many cases a factor of 2 in event distance can be accommodated by most runners. This doesn’t always work in detail. The athletics programme fits into this factor throughout the distance range from 100 m to the marathon. The exception, as pointed out by Martin Vlietstra is the 5000 m in the comments to [1]. There is then an argument for introducing the 3000 m into championship racing. In the USA, Daniel pointed out that High Schools compete over 1600 m and my recollection is that they also have 3200 m races, unique in the world. These are obvious conversions from the historical 1 mile and 2 mile races and have the advantage of the schools being able to retain their old records over the slightly longer distances. In Britain, our schools just accepted that their old records would be lost on metrication.

Back to the marathon, though, there was some discussion, particularly from Daniel in the comments to [1], regarding the shift from the official distance to a shorter marathon of 40 km. I think there are good reasons not to do this as the last 2 km of a marathon can be especially challenging, and this would not be the same for the last 2 km of a 40 km race. Some people argued for a rounding of the distance to 42 km. There is no particular reason to make it a round number, for the simple reason that roads have to be surveyed each time, so whether the distance is 42 200 or 42 000 m makes no difference to the surveying effort involved. The 5 m difference between 42 200 and the actual marathon distance is of course lost in the noise of the measurement. But athletics as a sport perhaps tends to be a bit of over rigorous at times. If the marathon distance had converted from imperial to metric with the answer 42 205 instead, the chances are that the 5 m would have got dropped. We are where we are, and the best approach is probably to leave it well alone at 42 195 m. Not 42.2 km, not 42 km, not 40 km. However you look at it, there is no obvious “creeping imperialism”, and the European athletics race walk distances were decided on to bring race walking more in alignment with running [7]. Half marathons and marathons are common on the continent, but 20 km and 40 km races are non-existent. Do you bring running into alignment with race walking, or the other way around?

Returning to the CG mile race, I wasn’t convinced that it was a successful experiment. One CG games record and an inaugural record is not that impressive. No doubt the weather in Glasgow had an impact on race tactics. But a 1.49 s improvement on the games record after 60 years of athletics performance is not in itself impressive. The women’s inaugural record, set in a heat, was a lot slower than Claudia’s 1500 m best for the year (worth about 12 s faster than her CG record over a mile) [6]. Coincidentally, Josh Kerr’s winning time was also about 12 s slower than his recent world record.

Acknowledgements: I’d like to thank the many people who provided discussion for the original article [1], with Daniel in particular making me aware of it [7].

References

[1] Commonwealth Games to replace the 1500 m by a mile race – Metric Views
[2] https://www.facebook.com/reel/27656037380685177
[3] Athletics at the 2026 Commonwealth Games – Men’s mile – Wikipedia
[4] Athletics at the 2026 Commonwealth Games – Women’s mile – Wikipedia
[5] Marathon – Wikipedia
[6] Claudia HOLLINGSWORTH | Profile | World Athletics
[7] https://worldathletics.org/download/download?filename=4428b6a0-ab13-429e-8a71-cf85d53976e0.pdf&urlslug=Decisions