Making sense of the controversial photo-finish at Amstel Gold
'What's the deal with that photo finish?', and other important questions.
'What's the deal with that photo finish?', and other important questions.
Sunday’s men’s Amstel Gold Race was one of the closest races in recent memory, with one of the most-talked-about outcomes. With a reduced peloton closing fast behind them, three riders were alone off the front: Wout van Aert (Jumbo-Visma), Tom Pidcock (Ineos Grenadiers) and Max Schachmann (Bora-Hansgrohe). At 200 metres remaining, Van Aert led out the sprint from the front, with Pidcock drawing level on his right hand side. To the finish they went, throwing their bikes to the line.
And then, the waiting began.
It took roughly 10 minutes for the race jury to scrutinise the photo finish and award the winner; roughly 10 minutes in which the TV footage of the finish was played and replayed in freeze-frame detail by broadcasters. That video suggested that Pidcock had won. However, after what felt like an eternity, the official result came through: Wout van Aert, by the most minuscule of margins. The Belgian had beaten the Brit by just 0.004 seconds.
Days later, the Amstel Gold finish continues to dominate discussion.
So how did we get here? Did the race jury get it wrong? Why do the TV pictures and finish-line photo show two different results? Who really won Amstel Gold?
Let’s get the ruler out and see how it all measures up.
The Amstel Gold Race finish was astonishingly close – almost historically close – but remarkably, you don’t have to go that far back to find another like it. On Wednesday at the women’s Brabantse Pijl, there was a near carbon-copy.
Ruth Winder (Trek-Segafredo) broke for the line on the right of the road, as Demi Vollering (SD Worx) went left. The Dutchwoman thought she had it, throwing one arm up in celebration as the US national champ threw her bike forward. The video camera suggested Vollering as the winner. The official finish line photo showed it was Winder’s victory. Two races in five days, each with two images apparently showing two different results.
I believe Vollering was even less than that. These lines are manually drawn with the software and the gap is calulated with the difference in the number of pixels between those two line here, but they didnt properly place it on Vollering’s front pic.twitter.com/gptKgHa7o2— La Flamme Rouge (@laflammerouge16) April 18, 2021
The nucleus at the centre of the swirl of confusion is the cameras: one a video camera, the other a photo finish camera.
After stage 3 of the 2016 Tour de France – which ended in a hard-fought sprint that saw Andre Greipel celebrate and Mark Cavendish eventually get awarded the win – VeloNews took a deep dive into how a photo finish camera works. There, tech writer Lennard Zinn explained that in those photo-finish images we see – with all of the riders together, a blurred background and wonky spokes – it’s not just one photo. The whole image is the finish line, shooting at 3,500 frames per second. The riders in the frame are shown not in the position they held when the first rider crossed the line, but the position they were in when they, themselves, crossed the line.
The X axis of the image is time, not distance. The frame is thousands of individual pictures of a fixed point, constructed into a larger whole.
That also explains why spokes appear curved in these photos, as parts moving faster are compressed horizontally and slower parts are stretched horizontally. As Zinn wrote in 2016, “the spoke at the bottom of the wheel is stationary since the wheel does not slip on the ground … since that spoke is stopped at the instant it passes bottom, dead center, it will get the most curvature, because as the film moves past, that spoke stays in the aperture for a longer period of time as it then moves further along around the wheel, thus appearing on the film for a longer period.
“The faster-moving spokes at the top are curved less, because they are moving faster and get past the aperture slit more quickly; their position over time on the photo is thus compressed horizontally because they have been in the light exposing the film for less time.”
If all that makes your mind feel as warped as those spokes, you’re not alone. Plenty of people have been left trying to grapple with the outcome of this year’s Amstel Gold Race, among them Tom Pidcock himself. In a (hastily deleted) tweet, Pidcock appeared to cast doubt on the result, saying that the photo finish “created more questions than have been answered”.
Pidcock retweeted an analysis from La Flamme Rouge which suggested that the photo finish camera had been set up “quite far from the entry of the black line”. This analysis was largely based around the advertising banners behind the riders – one yellow, advertising Amstel Radler (Ratebeer: “a pretty bad lemonade”) and one red, advertising Amstel Bier (Ratebeer: “a pretty gritty lager, objectively”).
La Flamme Rouge’s analysis centres around the absence of the red Amstel banner. If the photo-finish image is a strip of the finish replicated over and over, why doesn’t it appear in the background? The conclusion drawn is that the finish line camera was positioned before the finish line painted on the road.
However, a different angle suggests that the image should indeed have a yellow background rather than a red one. So what’s going on here?
The Dutch publication AD approached a UCI commissioner, Jempi Jooren – who did not officiate at Amstel Gold – for his perspective. Jooren, who is a specialist in the field of photo finishes, shed some light on the controversial finish, both from the perspective of an observer and a technical expert.
Asked by AD whether Wout van Aert was the true winner of Amstel Gold, Jooren said “after studying the photo finish, he was declared the winner. The regulations prescribe that the photo finish is leading for the result, if it cannot be seen with the naked eye.”
So why does the video make it look like Pidcock won?, the journalist asked. “I was not assigned to the Amstel and had the pleasure of seeing the race from the seat,” Jooren said. “I will tell you: I also thought based on the TV images that Pidcock was the winner. But now something important follows. The TV camera has a kind of wide-angle lens, which can give a distorted image. It’s a kind of optical deception.”
It’s only in extremely rare cases, Jooren says, that there is a disparity between what the TV camera and the photo-finish image shows – and doubly rare that this situation has happened twice in the space of less than a week. “This is an extreme situation. That it is so close almost never happens in practice,” Jooren told AD.
“Normally it is a nice addition for the viewer to see if someone wins with a half wheel or a whole wheel. You often do not see that well on the image of the front. Only in this specific case, when the truth is not properly portrayed, does it backfire.
“It happens that producers of television broadcasts get access to our photo finish, but that was apparently not the case in this case. It took a long time for the results to be official, so this image [from the TV still] is used. I don’t think that is illogical either, in 99 out of 100 cases you see the right winner on those images. Normally it is a nice service. Not now. I understand very well that many TV viewers and even the riders themselves were confused.”
It’s worth questioning, too, whether there’s a trust deficit in the UCI at this particular moment. High profile controversies – ranging from banned supertucks to discarded bottles to gnarly finish line crashes to dictator love-ins – have, rightly or wrongly, led many cycling fans to question whether the UCI has the competence to get things like a photo finish correct.
When this foundation is overlaid with a result that contradicts what viewers can see – or think they see – with their own eyes, you can understand how this stew of conspiracy, doubt and confusion has taken form.
The finish line photo is taken from a camera positioned on a truck above the finish line, and has the ability, Jooren says, “to observe down to a thousandth of a millimeter. A finish line at the race consists of a white area of 34 centimetres. Then a black ‘intermediate line’ of four centimetres and then again 34 white. The finish is where the first white strip ends and where the black strip starts.”
Both Jooren and the actual camera operator of the Amstel Gold finish, Renee Gielen, are adamant that there was no error. “We really checked whether it all went correctly,” Gielen told NOS, while Jooren reiterated to AD that the result was correct: “Certainly, 100%. That has been checked.”
That seems fairly unequivocal, but further doubt has arisen from the UCI’s technical guide for organisers of road events. Hidden away on pages 220-227 of a publication that is not best described as thrilling reading, the UCI outlines the organiser’s responsibilities for photo finishes. For WorldTour races, two cameras recording to two backup computers are aligned on opposite sides of the road, automatically capturing a minimum of 3,500 images per second, while timing must be accurate to 1/1000 of a second.
In the section titled ‘Alignment with the line’, the guide states that “The camera must be lined up with the white section just before the finish line in the direction of the race. The white background allows the commissaires to check the camera alignment. The camera, finish line and gantry must be perfectly aligned.” Accompanying images to demonstrate this show a line some distance from the black central line on the finish strip.
The UCI’s guidebook – or rather, the images accompanying it – appears to suggest that the positioning of the camera is at some whim of the operator, somewhere in the 34 centimetre expanse on the inner side of the black line.
However, the text and Jooren’s explanation, paired with a deeper understanding of the camera technology, makes it clear that the actual finish line is in the single pixel of white space immediately before the black line begins. As noted, the photo finish image that we see is not one image, but a composite of thousands of snapshots of a tiny, infallible sliver.
So is the finish line the white line or the black line in the middle of it? The answer, in fact, is that it is neither.
The black line is a visual approximation of the placement of the finish, but the image from the photo finish camera is the true arbiter of the result – not the other way around. It’s technical and it’s a bit confusing, but it’s clear-cut and the riders and officials are playing the same game with the same known set of rules.
According to UCI regulations, the photo finish verdict is final, because the finish line is what the photo finish says it is.
“The finish occurs at the instant that the tyre of the front wheel meets the vertical plane rising from the starting edge of the finishing line. To this end, the verdict of the photo finish shall be final,” says section 1.2.100 on page 41 of the UCI regulations.
That photo finish is not a potentially flawed visual representation of who won, but the unequivocal determinant of it. The rider who wins the photo finish, wins the race.
So: to return to Amstel Gold on Sunday, and the nailbiting moments leading up to its conclusion and the nailbiting moments immediately afterward. In the time between the end of the race and the verdict being handed down, viewers pored over a distorted video image that suggested Pidcock was the winner, without an understanding of the technical details that would ultimately determine the opposite. And it’s wild in that it has happened twice in a week.
Like Pidcock seems to have, many observers felt that the outcome ran counter to what they saw with their own eyes.
That’s unsatisfying, and it’s confusing. But it doesn’t change the outcome that was decided on Sunday afternoon: that Wout van Aert was the rightful winner of the 2021 men’s Amstel Gold Race.