Latest News

Share on twitter
Share on facebook

The conflicts at the heart of F1's TV graphics push

Years ago, the most that was offered were a few graphics detailing race positions, the gap between the leaders and a very rough estimate of how long pitstops took.

Nowadays, the rampant march of technology means that there is no end to statistics and data thrown out there regularly.

Thanks to F1’s tie-up with Amazon Web Services, viewers get battle forecasts, strategy predictions, car performance scores, corner analysis and tyre performance numbers.

But all this extra information has not come without its critics. Some have questioned the robustness of the data, with tyre performance graphics in particular having caused a stir a few years ago.

Furthermore, there has been some debate about whether or not the entertainment factor of following F1 races is ruined by graphics that predict when overtaking moves will happen (and their chance of success) or a likely order for the grid being rolled out ahead of the session.

The more accurate the graphics are in predicting what happens, the more that takes away from the spectacle. And if the graphics aren’t accurate, then what value do they bring?

F1 is well aware of such conflicts, and that it will never please all the people all the time.

However, hugely respected engineer Rob Smedley, F1’s director of data systems who is central to guiding the graphics, is adamant about how strong the data model is, and that its offerings are actually more spot on than many assume.

That was actually rammed home to him in the early stages of the AWS graphics appearing.

Rob Smedley

Rob Smedley

Photo by: Jean Petin / Motorsport Images

At the time, many fans had dismissed some of the information being given out, but F1 engineers were left impressed by just how close they were to reality.

“It was quite interesting actually, because when we first started this, there were a lot of comments of: just how can they do that, come on?” Smedley tells Autosport.

“But then what started to happen after these tyre performance graphics starting pinging up was, and obviously I’ve got lots of mates, and especially technical mates who work in the teams, I started getting quite a few messages saying ‘How did you do that then? Because when we’ve looked back on it, that’s pretty accurate!’

“Now it’s not right all the time, because it can’t be right all the time, as no model is correct 100 percent of the time, otherwise Formula 1 would be a fairly boring sport.

“But I thought it was just quite amusing. And it wasn’t just from one team either. It was from teams that we were showing where we thought their tyres were, and when they got the data off it and they’ve gone back and looked at it, they went ‘oh yeah, it’s not actually that bad is it?'”

While it is very easy for the data from the AWS graphics to be easily dismissed, Smedley says the processes used, and the quality of the information at the centre of it, are deep in their strength.

There are 300 sensors on each F1 car generating more than 1.1 million data points per second, plus F1 has access to the individual timing loops on each track – not just the three sectors that are widely used by teams and fans at home.

“The data set that we have here in one sense is really rich,” he said. “We’ve got 25 to 35 loops, depending on the track length. And you can glean a lot of information about that.

“If you think that teams have the three sectors, and they’re able to take a lot of competitor analytics from those three sectors, then if you then say I’ve got 25 loops, you’ve got an order of magnitude more of information.

“But like all data, once you get beyond that in the big data, you’ve got to know how to use it. It’s not a case of, I’ve got more data so therefore I’m better.

“Instead, that’s where the partnership with AWS comes in, because they’re big in data analytics and they’ve been doing this for however many decades.

“When you join the two companies together, where we know the data, we know the destination, we know how to use the data, we know what the destination is, but you need that big data analytics and that machine learning element to really give you that synergy and bump you up to the next level, that’s where I think it works.”

TV graphics, Braking performance

TV graphics, Braking performance

Photo by: Formula 1

It’s also a never-ending battle. For Smedley is clear that F1’s push for graphics is just like the development of cars, where one improvement simply opens the door for another one.

“I think it’s throwing up more and more questions. And it’s like if you go back 25 years, to when I first started, you get a little bit of data, because you think if I get that bit of data, then I’m going to know everything. But then all it does is just lead you to the next bit, and the next bit, and the next bit, and you never, ever stop.

“It’s just a constant mining of knowledge that you can never, ever stop, because once you get to the holy grail, you think if I could build something that would tell me all of that, that would really resolve all of my problems.

“You do that, and then you just realise that there’s more that you don’t know, and the more that you don’t know, it just gets bigger and bigger. That’s the exciting bit to be honest.”

But there are a host of external factors that needs to be balanced by F1 in picking out the important information, but not doing so in a way that it is overtly complex so alienates those watching at home.

“We’ve got to keep this thing at a level where it’s not information overload, where we’re not a bunch of engineers talking to a bunch of engineers, or a bunch of engineers talking to a bunch of data scientists,” continues Smedley.

“Then there would be a tiny demographic of people that are then enjoying F1 that are going to benefit from this.”

But perhaps the biggest tightrope act in delivering relevant graphics is in ensuring that they don’t ruin the fun for those watching the race.

If F1 produced graphics that 100% could accurately predict the grid on a Friday afternoon, and then throws out the right answers for pitstops, overtaking moves and the final result early in the race, then it would mean there was little point in watching the action.

Smedley is clear that the graphics need to enhance the experience for fans watching at home; not end up making watching the races redundant.

“We mustn’t ruin the jeopardy of the sport, because that’s why people tune in,” adds Smedley.

“If we all sat there on a Sunday morning and say, ‘right, we’ve got these brilliant models, here’s the finishing order,’ you’ve taken all the jeopardy out of it. You’ve told the story before it’s going to happen, and it’s quite tedious.

Sergio Perez, Red Bull Racing RB16B, Lewis Hamilton, Mercedes W12

Sergio Perez, Red Bull Racing RB16B, Lewis Hamilton, Mercedes W12

Photo by: Steven Tee / Motorsport Images

“What these graphics should do is say, there’s something going on in front of you now, but hang on a minute, because there’s something going to be happening in 20 laps’ time too.”

The message is clear that rather than AWS throwing out a predictor of when an overtaking move will happen as a spoiler, it’s actually an invite to entice people to stay tuned.

“F1 is so complex when you watch it,” says Smedley. “It’s got pitstops and tyre strategies going on, and it’s almost impossible to know everything that’s going on.

“It’s not like football where you watch it and what’s happening in front of you is it. [In F1], you’ve got to wait right to the very end sometimes for it all to unfold and the strategy to unfold.

“So what we’re trying to use these graphics for is to convey a lot of jeopardy, and to convey a lot of the stuff that’s coming up. Unless you’ve got these tools, it’s really, really difficult to tell that.

“You might be able to see that [Valtteri] Bottas is 15 seconds behind [Max] Verstappen, but in so many laps he’s going to catch him, and the overtaking probability is this. So actually what we’re trying to do is say there’s some action coming up, so stay tuned.

“It’s the same with the tyres. We can’t tell you what the outcome is going to be, because we just don’t have that knowledge.

“But what we can tell you is here’s the status of those tyres from the performance life, so more than likely, there’s a high probability there’s going to have to be a pitstop here which is going to cause some action, which is going to cause some further jeopardy.

“We must be responsible and use them for that purpose. I think if you go as far as explaining everything that’s going to happen and it always comes true, it’s somewhat spoiling the show.”



Related video

Related News
Related News