Hey, I scraped this data from the IAAF’s website [https://worldathletics.org/](https://worldathletics.org/), and used R to visualize
I can’t believe how universally the shoes were adopted. I guess it was really obvious the benefit they gave.
Ok but what happened between 2009 and 2013?
Clipping off all the times > 2;29 for the older era skews the trend line A LOT, to the point where the chart is arguably useless before 2010.
I don’t see it. If anything it looks more like a graph of when PEDs became easily available. Just like those men’s clinics that began operating in the same time period to sell men testosterone.
You would see a different curve if Nikes claims were to be believed. As it would be a constant change.
This also falls more in line with the investment into women’s sports making it more of an incentive to cheat.
The clipping of data does really make this less useful. But here’s my big question that I would find interesting to answer:
There’s a general improvement of times both in the Pre-Super Era (it appears to be about 2:29 in 1980 to 2:25 in 2018) as well as in the Super Era (looks like 2:24 in 2018/19 to 2:21 in 2025). So the Pre-Super Era saw a rough improvement of 0.10sec/year and the Super Era is seeing improvement of 0.42sec/year.
In theory time improvement in running is asymptotic, there is in fact a time that nobody will ever cross, but with the advent and widespread usage of Super Shoes has this just increased the rate at which we’ll reach a point of diminishing improvement or has it set a whole new bar of what the lower asymptote is?
It seems like nobody really has the answer to that since it’s such a new advent into the sport.
I love that the black line is labeled “Old Timey Sneakers” and goes right up to 2018.
Interesting data.
Questionable representation and conclusions.
As plotted, the OP (mis)represents that every athlete changed to a new class of shoe on the first of January, 2018. That seems…implausible.
The y-axis is truncated, and it appears that increasing numbers of runners and results are hidden as you go further back in time.
It’s not apparent that 2018 is the (or even a) breakpoint in the trend line. It looks like the times start to decline abruptly several years earlier, which brings the whole “supershoes” story into question.
(As an aside, since the chart only plots the top 100 runners in any year, it should be noted that the results can be skewed by changes in the size of the total pool of runners. If there are 10,000 marathoners, the top 100 is 1% of racers. If there are a million runners, the top 100 is the best 0.01%. If marathon running is getting more popular as a sport, looking at the top 100 runners means that you’re taking an increasingly elite subset.)
8 comments
Hey, I scraped this data from the IAAF’s website [https://worldathletics.org/](https://worldathletics.org/), and used R to visualize
I can’t believe how universally the shoes were adopted. I guess it was really obvious the benefit they gave.
Ok but what happened between 2009 and 2013?
Clipping off all the times > 2;29 for the older era skews the trend line A LOT, to the point where the chart is arguably useless before 2010.
I don’t see it. If anything it looks more like a graph of when PEDs became easily available. Just like those men’s clinics that began operating in the same time period to sell men testosterone.
You would see a different curve if Nikes claims were to be believed. As it would be a constant change.
This also falls more in line with the investment into women’s sports making it more of an incentive to cheat.
The clipping of data does really make this less useful. But here’s my big question that I would find interesting to answer:
There’s a general improvement of times both in the Pre-Super Era (it appears to be about 2:29 in 1980 to 2:25 in 2018) as well as in the Super Era (looks like 2:24 in 2018/19 to 2:21 in 2025). So the Pre-Super Era saw a rough improvement of 0.10sec/year and the Super Era is seeing improvement of 0.42sec/year.
In theory time improvement in running is asymptotic, there is in fact a time that nobody will ever cross, but with the advent and widespread usage of Super Shoes has this just increased the rate at which we’ll reach a point of diminishing improvement or has it set a whole new bar of what the lower asymptote is?
It seems like nobody really has the answer to that since it’s such a new advent into the sport.
I love that the black line is labeled “Old Timey Sneakers” and goes right up to 2018.
Interesting data.
Questionable representation and conclusions.
As plotted, the OP (mis)represents that every athlete changed to a new class of shoe on the first of January, 2018. That seems…implausible.
The y-axis is truncated, and it appears that increasing numbers of runners and results are hidden as you go further back in time.
It’s not apparent that 2018 is the (or even a) breakpoint in the trend line. It looks like the times start to decline abruptly several years earlier, which brings the whole “supershoes” story into question.
(As an aside, since the chart only plots the top 100 runners in any year, it should be noted that the results can be skewed by changes in the size of the total pool of runners. If there are 10,000 marathoners, the top 100 is 1% of racers. If there are a million runners, the top 100 is the best 0.01%. If marathon running is getting more popular as a sport, looking at the top 100 runners means that you’re taking an increasingly elite subset.)