Regression to the Mean: Searching for an explanation for "slow" times in the 2024 Diamond League Final

BRUSSELS, Belgium--The temperatures at the Diamond League Final were a stark departure from the summer heat of the Paris Games one month ago. The night time thermostat readings in Brussels were closer to the outside air temperatures in Glasgow, Scotland during the World Indoor Championships in early March. Fans in King Baudouin Stadium, some of them wearing scarfs and gloves, cited this as they questioned the perceived slowness of the performances on the track.

Was the weather to blame? Was the track surface inadequate? Had everyone peaked for the Olympics? Did too many fast runners skip the event? Were fan expectations, simply, too high? While spectators might have expected a continuation of the assault on the record books observed throughout the 2024 season, the likely scenario heading into Brussels was that there would be regression to the mean--following a single or a series of outlier performances, the following performance is likely to be ordinary, or, average.

In a sport where recent history has been flush with advancements in training, racing, and technology, everyone wants to believe that each excellent race represents a new echelon in performance. Though in events such as the women's 400m hurdles, times are certainly trending faster and the existence of a "new era" is indisputable, the overall result in Brussels proved to be a reset of expectations and a reality check for those submitting wild predictions.

In an outdoor season that began in April and featured 14 circuit meets, countless high-stakes national meets, and the Olympic games, it’s hard to normalize and compare performances. Instead, this article will compare this weekend's performances in Brussels against results from the preceding six Diamond League Final meets. The Diamond League switched to a winner-take-all finale format in 2017, so data from these end-of-season meets can be compared to determine whether "slow" means slow based on recent memory, or whether "slow" means slower than the average winning performance since 2017.

Methodology

For each track event, performances from the Diamond League finals from 2017 to 2024 have been retrieved and presented in a tabular format. The mean, or average, winning time each event has been computed as the DLF Mean: the Diamond League Final Mean. In the table, a column showing each year's winner's divergence from the mean is included. The fastest and slowest performances are highlighted using bold and italic print, respectively.

An in-stadium live take was captured before checking the historical performances. The idea behind this was to gather a subjective assessment of whether the winning time appeared to be fast or slow. That assessment then meets the hard data, and further context is provided regarding the performances and accompanying themes.

There was no final in 2020. Two-day same-location format debuted in 2021. For 2017 and 2019, the event was split between two different locations one week apart. In 2018, the event was held on consecutive days in two different locations, which made doubling impossible. In the current format, some doubles are possible while others are not (for example, the men's 1500m/5000m double wasn't possible). In the 2017/2019 format, most doubles generally were possible, and the qualify of performances may reflect that.

For brevity, field events have not been included. Sprinting performances have not been normalized for wind. Wind will not be addressed.

Men's 100m

Live take: Ackeem Blake's 9.93s was slow.

DLF Mean: 9.90

After reviewing the data: Blake's 9.93 was a bit slow in a final where the gold and silver medalists from Paris were not present. In a world where it takes sub 9.80 to generate meaningful debate, it's surprising to see so many winning times a tinge below ten seconds.

Women's 100m

Live take: Julien Alfred's 10.88 was slow.

DLF Mean: 10.81

After reviewing the data: Alfred's time was slow, and exceptionally slow compared to the prior three years of Jamaican dominance.

Men's 110m Hurdles

Live take: Sasha Zhoya's 13.16 was slow.

DLF Mean: 13.07

After reviewing the data: Zhoya's time was slow, nowhere near the 13.00 projected in a prediction article. Olympic champion Grant Holloway wasn't present, and the rest of the field ran sloppily.

Women's 100m Hurdles

Live take: Jasmine Camacho-Quinn's 12.38 was average.

DLF Mean: 12.43

After reviewing the data: Jasmine Camacho-Quinn ran very quickly to win the Diamond League Final. It must be mentioned that Tobi Amusan narrowly avoided a 4-year suspension for missing three consecutive drug tests and did not compete this season.

Men's 200m

Live take: Kenny Bednarek ran slightly faster than average

DLF Mean: 19.72

After reviewing the data: Aside from the 20 flat in 2017, the men's Diamond League 200m final has featured excellent times. Bednarek's performance was a challenge to the "slow track" narrative as it was indeed just slightly faster than the average. This came as two of his chief competitors, Noah Lyles and Erriyon Knighton, were not in the race.

Women's 200m

Live take: Brittany Brown's 200m win where she clocked 22.20 was about average.

DLF Mean: 21.87

After reviewing the data: Brown's 22.20 was the slowest time ever run in a Diamond League championship race. It was the same time she ran to earn bronze in Paris. In Brussels, she won comfortably in a bizarre final in which Julien Alfred and Gabby Thomas were absent, and Sha'Carri Richardson was a late scratch. 22.20 being a medal-winning time strongly suggests that event is in the midst of a regression to the mean.

Men's 400m

Live take: Charlie Dobson's 44.49 was slow.

DLF Mean: 44.35

After reviewing the data: In the shadow of Wayde van Niekerk's 43.03 world record in Rio de Janeiro, the Diamond League 400m finals haven't featured particularly impressive times. The only sub-44 second time came from Isaac Makwala in 2017. The slowest time came the next year, Fred Kerley's 44.80. Dobson's 44.49 is the second slowest.

The explanation here is that the 400m is a demanding event, and that none of the absolute best male athletes have strung together perfectly healthy seasons. The emerging pattern is that a random Diamond League race or that year's championship plays host to that year's most notable performance, and the rest of the season is left to slightly-inferior runners who are able to grind out races on the circuit.

Women's 400m

Live take: Mariledy Paulino's 49.45 was slow.

DLF Mean: 49.56

After reviewing the data: This was Meriledy Paulino's third consecutive Diamond League title. She dominated the regular season. Though the time was nothing revolutionary, she wasn't pushed to the line. Behind her, the fastest time was 50.32. In the "B" heat, which was not part of the Diamond League, Sydney McGlaughlin-Levrone ran 49.11 while running well ahead of the field. By the defined criteria, the final was not a slow race. However, if considering the performance of the whole field, the race was disappointingly slow.

Men's 400m Hurdles

Live take: Alison dos Santos's 47.93 was very slow.

DLF Mean: 47.39

After reviewing the data: The 2017 and 2018 finals were won by Kyron McMaster, who is only 27 years old and bagged sliver in Budapest last summer. When fans talk about a "big three" in the 400m hurdles, though, they aren't talking about McMaster, who has spent the last five years totally outclassed by Warstem Warholm, Alison dos Santos, and Rai Benjamin.

The precipitous drop in times was driven by the big three, and this year's enormous regression is due to two of the three being injured with the remaining competitor, dos Santos, being slightly diminished, having dropped out of his last race before Brussels. This was a slow race.

Women's 400m Hurdles

Live take: Femke Bol's 52.45 was slow.

DLF Mean: 52.98

After reviewing the data: This is one event that has, without a doubt, entered a new era since the format debuted. In 2017 and 2018, Dalilah Muhammed was good, but the new talent in the event, led by Sydney McGlaughlin-Levrone, brought the event to new heights.

This year's time seems slow only in comparison to Sydney's heroic world records. Femke Bol has won four consecutive Diamond League titles, as Sydney has avoided them all since she ran 52.85 as a 20-year-old in 2019. All that's being assessed her is Bol versus Bol, and compared to her past performances this was a totally average result.

Men's 800m

Live take: Wanyonyi's performance of 1:42.70 was faster than average.

DFL Mean: 1:43.62

After reviewing the data: 2017, 2018, 2021, and most of 2022 were down years for the men's 800m, where nobody came particularly close to David Rudisha's 2012 world record. It's widely known that the 800 has been in a slump since Rudisha's retirement. 1:42.70 is tied for the fastest ever in the DL final format.

This race was billed as a potential world record attempt, and Marco Arop took the race out on world record pace. Though it didn't precisely live up to its billing, the three-man race to the finish line was epic.

Women's 800m

Live take: Mary Moraa's time of 1:56.56 was about average but on the faster side.

DFL Mean: 1:56.93

After reviewing the data: Athing Mu's incredible 1:54.97 from 2023 is still fresh on the minds of American fans. Despite Mu not qualifying for Paris, and 1:54 woman Keeley Hodgkinson also being done for the season, Moraa did a great job of carrying on the tradition of fast times in the final.

It's worth noting that the final result average is artificially low, as the 2017 and 2018 1:55-range performances by Caster Semenya, an intersex athlete, are not considered fair by World Athletics' current standards.

Men's 1500m

Live take: Jakob's 3:30.37 was average, but felt slow.

DLF Mean: 3:30.33

After reviewing the data: Jakob and Timothy Cheriyuot ushered in an era of 1500m running that rewarded front-running instead of tactical races punctuated by fiery final laps. The 2017 DL final, won in 3:33.93, might have been a tactical fair, but nearly every top-tier 1500m race since the 2018 season has been a fast race, usually won in 3:30 or thereabouts.

Jakob, of course, deserves respect for this performance, as there's nothing easy about how he won the race. Based on Jakob's 2024 campaign, though, this 1500m likely didn't feel fast to him, either.

Women's 1500m

Live take: Faith Kipyegon's 3:54.75 was average.

DLF Mean: 3:56.69

After reviewing the data: Faith Kipyegon will break the world record at least once more before she retires or moves up to the 5k. She came into this Diamond League final as the class of the field, the infallible runner who quickly breaks up the field and wins in dominant fashion.

During this race, there was some inconsistent pacing from the rabbits, and so the majority of the field stuck with Faith late into the race. For this reason, her 3:54 high seemed slow. In actuality, it was well below the Diamond League Final mean.

Men's 3000m Steeplechase

Live take: Amos Serem's 8:06.90 was faster than average.

DLF Mean: 8:08.58

After reviewing the data: Aside from a very slow 2021 where the DL final was won in 8:17.45, there hasn't been much variance in the times in this event. Serem's triumph over El Bakkali came as a bit of a surprise. The other surprise here is that, in an event with an 7:52 world record, nobody has ever run until 8:00 in the final.

Women's 3000m Steeplechase

Live take: Faith Cherotich's 9:02.36 was slower than average.

DLF Mean: 8:59.43

After reviewing the data: Amazingly, even after Winfred Yavi pumped the brakes and dared her foes to pass her, the women's 3k ended up just slightly slower than the mean. The early pace was hot but the race slowed for an inconvenience for Winfred Yavi, and a bit of an anticlimax for fans expecting to see the runners stay with the pace lights.

Men's 5000m

Live take: Berihu Aregawi's 12:43.66 was about average.

DLF Mean: 12:53.02

After reviewing the data: We've been spoiled by fast distance times lately. Though the 5k on the track doesn't always end up being an all-out affair, 12:43 is all-out, and the target was likely sub 12:40. There was nothing average about Aregawi's time, this is the type of performance that would have been drooled over a decade ago.

Women's 5000m

Live take: Beatrice Chebet's 14:09 was much faster than average. This appeared to be the highest-effort and most dominant performance of the track meet.

DLF Mean: 14:23.08

After reviewing the data: Chebet's time was the second-fastest ever run in a Diamond League Final, behind the 14:00.21 world record from last year. Her performance also would rank her 4th on the all-time list, if she wasn't already ranked 3rd. Her courageous solo effort was a highlight of the meet.

Conclusion

The influence of the excellent times witnessed during the 2024 Olympic year cannot be ignored. To whatever extent the Brussels Diamond League might not have met fan expectations, 10 of 18 performances were faster than the Diamond League Final Mean, calculated since 2017.

Compared to Olympic performances and some of the breakthrough marks set during the regular season, there was some regression here. At the the least, the Memorial Van Damme Meet was not the superlative finale in the same way that the Olympics is. There were no true outlier performances.

Hopefully, this analysis helped to exhibit that, as any sample size grows, outlier performances will continue to be paired with mediocre ones. The meet was not "slow" compared to past Diamond League championships; it was, in a word, average.