Three Women and the 6.706 km Loop: What the Data Has Not Yet Recorded at the Rasselbock Backyard Ultra

Every hour, exactly once, a whistle sounds at Hardwick Estate in Derbyshire. 6.706 km. No more, no less. Runners must complete the loop and be back at the start line before the next whistle. Anyone absent when the new hour begins stops there. There is no conventional finish line, no silver medal, no official third place.

By the end of the race, three people were still standing at the start line. All three were women.

That is the entirety of the result the original report left behind. No names. No total distance. No loop count. No finishing time for the final lap. Not a single measurable number describing the ability of those three women, while a substantial share of the article's word count went to the history of a sixteenth-century hall.

I read that report three times, and only on the third pass did I realise the most interesting thing sat somewhere quite different from the headline.

Context: a format with no textbook

Backyard ultra is not a marquee event. It is a community-grown branch of ultra-distance running, sitting outside World Athletics governance, with no qualifying standard, no national ranking table, no ranking points, and almost certainly no mandatory doping testing at this tier.

The format originated with Gary Cantrell, known in US running circles as Lazarus Lake, through Big's Backyard Ultra in Tennessee from 2026. The principle is simple and brutal: a fixed loop, a start on the hour every hour, run until only one person remains. The last one standing is the only officially recognised finisher. Everyone else, including the runner who lasted second longest, is recorded as DNF, meaning Did Not Finish.

This is where many sports readers misread the story. In a backyard ultra there is no conventional 1-2-3 table. Only one person finishes. But most races still record a group of the last three standing, as a community convention, and media call it a podium. What the headline "first women-only podium" actually refers to is that convention, not an official three-step podium.

Rasselbock Backyard Ultra is staged by a private race organiser of the same name, at Hardwick Estate in Derbyshire, a property managed by the National Trust. The loop runs 4.16 miles, equal to 6.706 km. Sustained across 24 hours, 24 laps equal 100 miles, roughly 160.9 km.

The only person named in the entire report is Matt Dillon, a representative of Hardwick Hall. He is not an athlete. He is not a coach. He is a venue spokesperson. Much of the story circles Bess of Hardwick, that is Elizabeth Talbot, Countess of Shrewsbury in the sixteenth century, the most powerful woman in England after Queen Elizabeth I in her era, and the figure most tied to the building of Hardwick Hall.

When the most fully identified figure in a sports report is a venue spokesperson while three athletes go unnamed, you already know which angle the piece was written from.

The core: the one hard dataset worth analysing

Across the whole report there is exactly one hard dataset, and it is not about the three women.

It is the size and gender composition of the field: 141 registered runners, roughly 30 women and roughly 111 men. That is 21.3% and 78.7%, or about one woman for every 3.7 men.

That number is dry, but it allows a calculation the original report never performed. If the last three standing were determined purely at random from 141 people, regardless of gender and ability, the probability that all three would be women is the number of ways to pick 3 women from 30 divided by the number of ways to pick any 3 from 141.

Choosing 3 women from 30 gives 4,060 combinations. Choosing 3 people from 141 gives 457,310. The ratio is 0.00888, or about 0.89%, roughly 1 in 113.

I stress this: it is a naive random model and it is wrong in sporting terms. It assumes every athlete shares the same ability distribution, whereas in reality the female cohort entering backyard ultras is self-selected, with average commitment above the general field. Adjust for true ability distribution and the real probability rises above 0.89%. But even multiplied three or four times over, this remains a result that deviates from statistical expectation in a notable way.

And this is where the headline starts missing its own centre of gravity. The newsworthy event is not the "women-only podium". The newsworthy event is that 21% of the field filled all three of the last-standing places. The outcome far outpaced the representation, and that is the dataset worth tracking going forward.

What this format actually measures

The mandated pace is 6.706 km within 60 minutes, roughly 8 minutes 57 seconds per km, or about 14 minutes 24 seconds per mile. To anyone who runs seriously, that is comfortable. It sits in the aerobic zone, below lactate threshold for most consistent trainers. You do not need to be fast to complete one lap.

You need something else to complete 24.

The determining variables fall into four groups. First, pacing discipline: newcomers often run the opening lap too fast because it feels light, then pay for it at lap fifteen while waiting at the start line with a cooled-down body. Second, sleep-deprivation tolerance, because after hour twenty the central nervous system begins degrading decision-making, and the only decision required each hour happens to be the most important one. Third, crew quality, the personal support team handling nutrition, shoe changes, blister care, warmth, and sometimes just a single sentence. Fourth, management of accumulated micro-trauma: plantar fasciitis, bone stress reactions, Achilles tendinopathy, hip joint pain, none of which appear in a single 42 km run but all of which appear after eighteen hours of repetitive loading.

Not one of those four groups is a variable where male physiology confers a decisive advantage.

This is the most important analytical point in the piece, and I want to state it plainly: the backyard ultra format strips away almost every factor on which the gender gap in athletics is built. There is no finishing kick. No acceleration. No decisive surge. No explosive strength component, no muscle-mass advantage, no moment where peak power is summoned across three seconds. All that remains is sustained performance at low intensity over a very long duration, and at that length the gender gap compresses considerably compared with 800 m or the marathon.

In other words: an all-female podium at a backyard ultra may not be evidence that women are catching men in running generally. It may simply be evidence that this format does not hand men the advantage other distances still do. Those are very different conclusions, and media tend to pick the first because it sounds better.

I still have to state the rest of the physiology. Repeated loading over more than twenty hours is not a gender-neutral game. Bone density, endocrinology, menstrual-cycle variation, repetitive-load injury rates, and how a body handles prolonged sleep loss can all create differences in either direction. But public data on those points at grassroots level is close to zero. I have no table to cross-check against, so I draw no conclusion.

I worship data, but I pray through real-world verification.

The contrarian angle: a record with no registry

The headline says this was the first women-only podium at a UK backyard ultra.

Who says so? The race organisers.

Is there independent confirmation? None in the source.

Is there a national register of UK backyard ultra podium compositions to check against? To my knowledge, no such public register exists.

Backyard ultra has no unified global governing body. No federation. No record-ratification system. No adjudication panel. Every "first ever" claim in this discipline depends on whether the community and organisers maintain a collective memory. If a UK backyard ultra had previously produced an all-female top three, and nobody recorded it, the claim is false. If none had, the claim is true. Neither possibility can be settled from public data.

Before I believe a reputation, I need to see the data behind it.

Here I see no data. I see a statement attributed to organisers, and a void behind it.

A sample size of one

There is a bigger problem than the claim's verifiability.

It is the denominator.

One race. One weekend. Three unnamed athletes. No loop count, no total distance, no time. If someone asked me whether British women are dominating ultra-distance racing, I could not answer with this dataset. I could not even say how far those three women ran.

A season should be read as a sequence of probabilities, not a sequence of events. Here we do not even have a sequence. We have a single data point, and that point is missing its vertical axis.

In my trade there is a familiar trap: taking one victory and inferring a trend. In 2026 I published a series using xGA to show that the CLB Thanh Hoa defence, widely praised as the best in V.League, was actually conceding more than expected, with 1.9 xGA per match and a goalkeeper save rate of just 64%. The coaching staff called me the guy sitting in the cold room. On 7 February 2026, Thanh Hoa lost 0-3 to Ulsan Hyundai in the AFC Champions League play-off, exactly the script the data had indicated.

I retell that not to praise myself but to say the opposite: I could publish that conclusion because I had twelve rounds of data, an xGA table, a save-rate figure, and match-by-match cross-checks. With only one match, I would have written nothing.

At Rasselbock, nobody has twelve rounds of data. Nobody has a table. The report supplies no name, time, or lap count for any of the three women.

What cannot be measured should not be written. But that does not mean total silence. It means writing the measurable part correctly, and stating clearly which part is not.

The measurable part here is 21%.

A second layer the report reveals by accident

There is a detail I kept returning to. Bess of Hardwick occupies a large share of the article's words, far more than the race result. Matt Dillon, the Hardwick Hall representative, is the only person quoted verbatim. The 6.706 km loop is mentioned, but no analytical table accompanies it.

Put those fragments together and the structure becomes fairly clear: this is a piece built around heritage and venue tourism as much as around sport. A National Trust property, a sixteenth-century countess famous for power and for building Hardwick Hall, a grassroots ultra drawing 141 runners, and three women who lasted to the end. Every piece fits into a story about women and heritage.

I am not saying that is wrong. I am saying it is a business model worth tracking. Grassroots races paired with heritage venues are becoming a two-way sponsorship channel: organisers gain a beautiful site and legitimacy, venues gain footfall and free media. In the UK the model is expanding. In Vietnam, trail and ultra races tied to conservation areas, national parks and heritage sites are following a similar trajectory.

But if your goal is to assess sporting ability, that structure is a warning. When a report's centre of gravity sits at the venue, the competitive data is usually the first thing cut. And when competitive data is cut, the result becomes far harder to verify than its surface suggests.

My two-hat principle

This situation revives an ethical question I have stood inside. In 2026 I was simultaneously writing public analysis and serving as a data consultant to a club. When that club struggled, I had to choose: disclose internal data to keep my journalist role, or stay silent to protect the team. I chose the team, and my former newsroom severed ties with me.

Since then I have held a two-hat principle: never mix a club's proprietary data into public writing, and use only data from official platforms.

That principle applies directly to Rasselbock. I have no access to the race timing data. I have no lap table. So I may analyse only the public data: 141 runners, 21% women, 6.706 km per hour, and one organiser statement. Anything beyond those four points I must label as inference, not fact.

Signals for the next cycle

If you follow this field, there are four indicators worth recording.

First, names and lap counts. When a backyard ultra ends, the most verifiable result is the list of runners who reached the final lap with each person's lap count. If the next report carries that data, we can assess real ability. If it still does not, that itself signals the event's level of professionalisation.

Second, verification of the "first ever" claim through the UK backyard ultra community. If someone contradicts or independently confirms it, the story's value shifts in either direction.

Third, and most important to me, the female share in subsequent races. One race's 21% says nothing. 21% sustained across three or four seasons is a trend. From my experience tracking matches and athletics events, representation indicators always run a few years ahead of performance indicators. If the female share in UK ultra-distance races holds above 21% across multiple seasons, an all-female podium becomes ordinary rather than remarkable.

Fourth, the heritage-venue plus grassroots-race model. If conservation areas and heritage sites keep opening up to races, that becomes a genuine growth channel for the industry, and it will reshape how small races build their cost structures.

In Vietnam I have no official figures on the female share in trail and ultra races. That is a gap worth filling, and I have started building my own tracking sheet for next season, the same way I once built my own PPDA spreadsheet rather than waiting for official data to appear. With empty stands, I heard what twenty thousand people used to drown out: data.

Those three women at Hardwick Estate may be a full stop, or a question mark. They outlasted 138 others in a format where the only prize goes to the last survivor. They were not named.

If the organisers publish each runner's lap count, we will know whether this was a historic moment or a fine weekend in Derbyshire. Until then, the only thing I will assert is the number 21% standing beside the number 100%.

Luck is the residual my model cannot explain, and I never round it to zero. But a residual is not evidence either. It is only a place to put the next question.