Super Shoes and the Interrogation of Marathon Records

I remember the morning of October 8, 2026. Chicago. The temperature was around ten degrees Celsius, the wind was light, and the course was as flat as a sheet of paper stretched along Lake Michigan. Kelvin Kiptum, twenty-three years old, from Kenya, crossed the line of the 42.195 km race in 2 hours, 00 minutes, 35 seconds. The men's marathon world record was erased from the books on a morning when most Vietnamese viewers were still asleep.

My phone kept buzzing. Colleagues asked. Editors asked. Ordinary people who watch athletics only once every four years asked. All with the same question: is this real?

I did not answer right away. I opened the charts.

That is how I have worked for twenty-seven years. Not shouting about whether a record is real or fake. Just calling each number to testify, letting them confront one another.

Data never argues. It only exposes the truth.

And in Kiptum's case, the data said something the naked eye could not see: this record came from structure, not from inspiration.

Six Years of Rewriting History

To understand why, we must go back six years.

In October 2026, also in Chicago, Brigid Kosgei ran 2 hours 14 minutes 04 seconds, breaking Paula Radcliffe's women's marathon world record that had stood since 2026. A week earlier, in Vienna, Eliud Kipchoge completed a marathon under two hours in the INEOS 1:59 project, with 1 hour 59 minutes 40 seconds. That performance was not recognized as an official record, because of the pacing group, the wind-blocking formation, and the laser-guided pacing strip. But it marked a psychological boundary: two hours was no longer impossible.

Both 2026 events were tied to a variable the old charts had never contained: the shoe.

Nike launched the Vaporfly line with a carbon plate and supercritical foam, then the Alphafly. Other brands rushed in. Adidas had the Adizero Adios Pro. Asics had the Metaspeed. Puma had the Deviate Nitro. A technology race erupted not on the track but on the design table.

In January 2026, World Athletics issued its first set of regulations on competition shoes. The stack height limit for road running was 40 mm, with only one carbon plate or an equivalent rigid structure permitted. For track shoes, 25 mm; for spikes, 20 mm. Just as important, every shoe model had to be available on the open retail market for at least four months before an athlete competed in it.

This was a compromise regulation. It did not ban technology; it only framed it. Legally, it made sense. Philosophically, it raised a question with no definitive answer: when is a record a human achievement, and when is it a laboratory achievement?

I was present at press conferences around that time in various roles. What I remember most is not the numbers in the rules but the silence of coaches when asked about the threshold limits. No one wanted to say a simple thing: every marathon record had long been a product of materials science; it was just that the science used to be invisible.

Calling Each Number to Testify

Now let the data work.

Each number is a testimony. I only carry out the interrogation.

The starting point is the curve of the men's marathon record. Before 2026, the world record progressed slowly. Paul Tergat ran 2:04:55 in Berlin in 2026. Haile Gebrselassie ran 2:04:26 in Berlin in 2026, then 2:03:59 in Berlin in 2026. Patrick Makau ran 2:03:38 in Berlin in 2026. Wilson Kipsang ran 2:03:23 in Berlin in 2026. Dennis Kimetto ran 2:02:57 in Berlin in 2026. On average, each step forward took years and tens of seconds.

Then came Eliud Kipchoge. Berlin 2026: 2:01:39. Berlin 2026: 2:01:09. Then Kelvin Kiptum. Valencia 2026, his first official marathon: 2:01:53. London 2026: 2:01:25. Chicago 2026: 2:00:35.

The gaps between the markers reveal something alarming. Kimetto took four years to go from 2:02:57 down to 2:01:39, a gain of 78 seconds. Kiptum went from 2:01:25 down to 2:00:35 in under six months, a gain of 50 seconds, at twenty-three years old, in only his third marathon.

That is the point where any analyst must stop. A leap like that, under normal conditions, is a red flag. But the conditions here were not normal, and I needed to test each variable before concluding.

The first variable is the course. Chicago is famous for being flat, with few turns and an elevation close to sea level. These are ideal conditions, but they are not exclusive to Kiptum. Berlin is also flat, and Kipchoge ran there.

The second variable is the weather. Chicago's temperature that day was around ten degrees, nearly perfect for a marathon. No harsh sun, no strong wind. This is a benefit the raw chart does not fully capture.

The third variable is the pacers. Kiptum ran most of the distance with quality pacers, which helped him hold a steady rhythm around 2:51 per kilometer. Without pacers, this leap would be hard to achieve.

The fourth variable is the shoe. Kiptum wore a thick-soled shoe with a carbon plate, legal under the 40 mm rule. This is the most important and most controversial variable.

So how should we read Kiptum's leap?

The answer lies in separating two things the media often conflates: biological potential and the technology dividend. If part of the advance came from the shoe, then that part does not belong to Kiptum as an individual. But it also does not belong to Kipchoge, nor to Kimetto. An entire generation of marathoners is running on a new type of sole.

In other words, we are not comparing Kiptum with Kipchoge. We are comparing a new shoe era with an old shoe era, then attributing the difference to the individual who ran last.

There is a simpler test. If the shoe were the sole cause, then every top athlete would improve in unison. In reality, that did not happen. The number of runners breaking 2:05 increased, but the rate of increase was uneven. Some advanced dramatically; others stood still. That shows the shoe is a necessary condition, not a sufficient one.

So which part belongs to Kiptum?

Here I must say something many colleagues hesitate to say: Kiptum was a rare biological phenomenon. He had an ideal height, a long stride, and most importantly, a capacity to tolerate pain at the lactate threshold that very few possess. Based on my years of experience watching athletics, athletes who break through at twenty-three usually possess two things at once: a young physical foundation and the tactical composure of a veteran. Kiptum had both.

But I will not conclude. Because the most important data about Kiptum is not the 2:00:35 figure. It is the figure that should have appeared afterward, and never will.

On February 11, 2026, Kelvin Kiptum died in a car accident in Kenya, at just twenty-four years old. With him, a data curve was cut off. We will never know how the Kiptum of twenty-five, twenty-seven, or thirty would have run. Every comparison with Kipchoge from now on is a comparison with a man who no longer has a second chance.

That is why I am always cautious about steering the story toward records. Every record is a testimony, and every testimony is cut off by the conditions of its taking. Kiptum testified once, then fell silent forever.

The Other Side of the Same Mirror

Data analysis has a blind spot I must name.

When Kiptum ran 2:00:35, most questions revolved around the shoe, the pacers, the course. When Ruth Chepngetich ran 2 hours 09 minutes 56 seconds in Chicago in 2026, the first time a woman broke the 2:10 barrier, the questions were far heavier. Suspicion. The language on forums was no longer technical doubt; it was accusation.

I am not here to defend or accuse anyone. I am only reading the asymmetry.

The same regulatory system, the same shoe generation, the same Chicago course. When a man breaks a record, the charts are called up to explain the performance. When a woman breaks a record, the charts are called up to doubt the performance. The charts here no longer serve to explain. They are used to convict, and the verdict always precedes the evidence.

People may laugh at my name, but they cannot laugh at my charts.

And charts do not distinguish by gender. They distinguish only between evidence and speculation.

I was once mocked for my gender. In August 2026, when I analyzed the midfield gaps in a 4-2-3-1 formation in a Chinese national league match, a social media account with over half a million followers publicly asked what a woman could possibly understand about tactics. I did not argue. I reviewed six matches, tallied the passing rate in the central zone, and wrote a two-thousand-word rebuttal. It was shared more than eight thousand times.

That experience taught me one thing: when doubted, do not defend. Let the data stand on its own.

About Those Without Pacers

Here I want to tell a story that the 2:00:35 record has overshadowed.

Elite marathon running does not belong only to Kenyans and Ethiopians. It also includes athletes from countries without professional pacing systems, without altitude training centers, without the budget for the newest thick-soled shoes. They run alone, at smaller meets, with times rarely mentioned.

The same 40 mm rule applies, but the price of a top-tier shoe equals several months of income in some countries. The rule makes technology free on paper, but not free in reality.

This is where the shoe rule, though technically reasonable, becomes invisible on the question of fairness. An athlete running 2:08 in a country without shoe sponsorship is running faster, in purely physical terms, than an athlete running 2:04 in the best shoe on the market.

The media does not report on the 2:08 runner. He has no record, no glory, no question about his shoes. But he is the one closest to the human limit.

I call him the underdog. And my job is to pull him into the light with numbers.

The Counterintuitive Angle

If you have read this far and are still waiting for a verdict on Kiptum, I must disappoint you.

At the start of this article, I promised to call the data to testify, not to convict. But there is one counterintuitive conclusion I am willing to offer: the debate over super shoes, six years on, still has not answered the right question.

People ask: are shoes technological doping?

The right question should be: does the record system still have meaning when the technological variable changes faster than the biological one?

This is the crux. A record only has meaning when two athletes at two different points in time can be compared relatively fairly. When technology changes between generations, the comparison table loses its stability. The 2:00:35 record does not lie, but it also does not tell the whole story. It only tells the truth of 2026, with the technology of 2026, on the course of 2026.

Another counterintuitive point: tightening technology rules may slow records down, but it also makes them more credible. If World Athletics continues to freeze the 40 mm limit, in the next ten years we may see fewer records, but each record will carry more weight. That is a trade-off the media rarely dares to state, because fewer records means less news, fewer views, less controversy.

And there is one more blind spot. We spend thousands of hours analyzing shoe soles, yet almost no comparable time analyzing unequal competition conditions: temperature, altitude, pacer quality, race scheduling. The shoe is the easiest variable to measure, so it inevitably becomes the most blamed variable. Data does not choose its own focus. People choose.

The Next Curve

In the coming season, there are three curves I will be watching.

The first curve is the women's marathon record. From Kosgei's 2:14:04 in 2026 to Chepngetich's 2:09:56 in 2026 is a leap even larger than Kiptum's. If this leap is repeated by another athlete, we have moved to a new standard. If it stands still, that is a signal to revisit the entire old chart.

The second curve is World Athletics' stance on the shoe stack limit. Every time the body nudges one millimeter, athletics history splits into before and after.

The third curve, and the most important, is the runners breaking 2:08 in countries without support systems. They are testimonies that have not yet been interrogated.

I have been wrong many times in my career. Once I mispronounced a player's name three times in a single half, and it took me three weeks of reviewing the entire dossier to correct it. Since then, I have built a five-step process for every analysis: check the data, check the source, check the context, check the cross-reference, and only then am I allowed to conclude.

Process is not a cage. It is the shell that protects freedom.

With Kiptum, I will not conclude. I will hold onto the number and wait for the next curve. Because data never argues, but it is also never in a hurry.

History does not repeat, but it echoes.