Spectators still find tyre graphic difficult to understand – Pirellli

2019 F1 season

Posted on

| Written by and

Pirelli believes Formula 1 fans are still confused about the new tyre graphic added to broadcasts this year.

The Tyre Performance graphic, which was introduced during the Japanese Grand Prix, shows percentage values describing the performance of a driver’s tyres. It is produced by F1 and created using publicly-available data, not information supplied by Pirelli.

The manufacturer’s head of F1 and car racing Mario Isola is concerned the graphic still isn’t fully understood by fans.

“What they want to show is the performance life of the tyre,” he said. “And the performance life of the tyre is an interesting number because the idea is to give the spectator the feeling [whether] a car can fight with another car, what [might] happen in a few laps’ [time]?

“The performance life of the tyre is also influenced by the pace management. So if for any reason a driver that has 30 percent of performance life in the tyre realises that he will have degradation of the tyre in a few laps he can drive the car in a way that reduces this degradation.

“That means that the 30 percent stay can stay as 30 percent for a number of laps. That’s probably the most difficult part to explain to spectators.

“Everybody has in mind that you start from one hundred percent and you finish at zero percent. But it’s not linear.”

Isols believes fans expect the a drivers’ ‘tyre performance’ value would fall every lap. “But the reality is that this curve can be different if you are in free air, if you follow another car, if you are pushing with the level of fuel, with the energy that you’re putting into the tyre.”

F1 said it made improvements to the graphic after Pirelli described it as “misleading” following the Suzuka race. However Isola believes it needs to be simplified further.

“If you remember the first time at Suzuka we had the graphic on the screen, it was a bit of a surprise because in our understanding it was related to the wear life. Then we had meetings where they explained the system and it is clear that is related to the performance life.

“The next step is to understand how to give this information in a way that is easy, on one side and also is telling you exactly what is going to happen.”

Advert | Become a RaceFans supporter and go ad-free

2019 F1 season

Browse all 2019 F1 season articles

44 comments on “Spectators still find tyre graphic difficult to understand – Pirellli”

  1. Why is it that everyone involved in F1 thinks we are stupid?

    1. +1
      Personally, when I see that graph, what i’m wondering is if I can trust the data or if it’s a gimmick giving a very high level idea of where the tires are… Not how to read it…

      1. Personally, when I see that graph, what i’m wondering is if I can trust the data or if it’s a gimmick giving a very high level idea of where the tires are… Not how to read it…

        Exactly! They say they’re deriving the stat from “publicly available data,” where can I get access to this data? I’d like to see the specifics and draw my own conclusions.

      2. I thought this too and asked that very question to the Mercedes F1 guys at big data London. He said it was basically guess work. Probably based on ml that is currently and probably never will be accurate as to when hammer time begins or as to how much a car actually slid around a corner.

    2. @mrboerns they even think that no one will notice how stupid it is to keep Singapore 08 result

  2. “…also is telling you exactly what is going to happen.”

    What a bunch of ….. How can anyone involved in the sport say this with a straight face and not understand the sheer ridiculousness of it.

    I despair.

  3. Credit to FOM and their never-ending quest to remove all mystery and anticipation from the sport. Just give me the time deltas. I’ll use my eyes and brain to work out the rest, thanks.

  4. It’s not difficult at all to understand. It’s just plain wrong most of the time.

    One race, I forget which, the graphic showed Hamilton had 70% left on 2 of his tyres after 21 laps. 2 laps later, the graphic showed he suddenly had 40% left, and he wasn’t driving like a taxi driver all of a sudden.

    It’s pointless if it’s not accurate – rather just not show it. Either the first one or the last one was wrong. And this is just one example – either the first one or the last one was just a pure thumb-suck.

    1. @ho3n3r And what is your expectation of the lap time difference between 70% and 40% tires performance? What are you basing your expectation from?

      It seems you misinterpreting what the graphics try to communicate, although to be fair as it is now we don’t have enough info told to us to accurately translate the tire performance value into actual measurable data (eg. laptime).

  5. Can they also please tell us what the driver had for breakfast, their opinions on climate issues, and the brand of razor they use. The public has a right to know! :-)

    1. Razor’s … where have you been.?
      Haven’t seen a clean shaven driver on race-day for quite a while. Any that were over 18.
      Even C. Horner has taken to “The Stubble” for race weekends.
      Don’t know about anyone else, but if I miss a day (or two) it can be downright painful to catch up.
      Maybe this is a question for Dieter.? Why the (NHL hockey terminology here) Playoff Beard.?
      My guess is that it helps with the breathing and cooling of the drivers while wearing the required balaclava.
      Yes, it looks like this could be classified as a “driver aid” or a “move-able aero device”. Need to ban it.

  6. I quite like the idea.
    Have major doubts about the accuracy though.

    1. I actually like how Pirelli digs into F1/Liberty to get to work on this graphic / idea to make it actually clear what it means and at least somewhat more accurate @coldfly.

      Off course they do it out of self interest – i am sure people were already complaining about Pirelli not knowing their business with how wonky the predictions wer, expecting the info to come from the manufacturer – but at least they have the information and the interest+connections to stay on it until it works better.

    2. There could be some sensors, and AI analisys and involved, that would evolve with every case seen.
      Visual sensors on the sidepods for example to take pictures of the tyre surface.
      And some thermal sensors also.
      Variables: laps, fuel load, pace and temperature history, track surface, driving style, well done or overdone outlap.
      As IoT and AI are so much trending, it’s super possible to be done much better.

      1. Also, as I see talks about publicly acailable data. So if the visualization and prediction is ultra accurate, then maybe teams maybe won’t be happy giving away so much information to opponents realtime. If you have all history and realtime data too, then strategy decisions are computable on the fly at a next level. It would be nice, but maybe team’s won’t like it.

        This is similar to online poker. I had been a micro + lowstakes nightmare at top tier rooms for years, sadly i joined too late to get rich, but it’s still a cool feat as I’m from a poor country, as initial conditions are much worse than a really first world player’s. I have heard proven highstakes guys talking about how much levels like NL1000$ is beaten by bots 5 years before.
        AI is capable of inventing nonsense but still working styles and patterns, these can send opponents to hell. I think now all levels are infested at online poker, so i quit since years and it was so easy to erase from memories, despite of i’m still quite good.

  7. In Brazil, Hamilton had 80/70% left and Verstappen 40/30% at the end of the stint. Yet Hamilton pitted a lap earlier. The same was shown in Austin iirc. 70% tire life left after a long stint and the next lap Lewis pits.

    The graphic is utter pointless. Showing tires + laps on them in the tower and you can work it out yourself.

    1. In the case of Brazil it might be true though. Hamilton went for the undercut because that really looked like their only chance of getting ahead of the faster Red Bull. Going long doesn’t work against a faster car that’s just pacing itself to keep a gap.

      So Hamilton would indeed have stopped well before his tyres would have been done.

  8. I’m honestly curious about which data/algorithm they use to get to that X% number. Teams can monitor pressure/temperature, and still they put a rotary switch in the steering wheels to get feedback from the drivers of what phase the tyre is in. Only after they’ve change the tyres the team measures the tread and can get information about how marginal they were on tyre wear.

    1. @warheart Prediction model by AI trained with publicly available data. It’s been stated few times before and I think there’s an article not long ago. Simply put, AWS try to make and AI that predict using what we as viewers has (theoretically) access to. Remember how AI can predict someone pregnant and start serving proper ads just by analyzing their browsing habit, without having data on pregnancy test or their copulation schedule? Same thing here except the browsing habits is publicly available data and pregnancy test is the telemetry that private to the teams only.

      1. @sonicslv “Prediction model by AI trained with publicly available data” is AWS’ equivalente of any driver’s “it’s a challenging track and we’ll do our best to be competitive” PR talk. Yeah, it makes sense but it doesn’t give any information. Are they infering tyre degradation from lap times? From deltas to personal best lap? To “train” an AI you need first and foremost reliable data, and AFAIK there’s no reliable data regarding tyres publicly available. You can’t put a graphic saying a tyre is at “70%” and not say what that 70% is of. There’s a great comment by @leonardodicappucino pointing this.

        1. @warheart Your question is what the data/algorithm they used. And it’s not PR talk, at least if you have understanding on how programming an AI using big data analysis and prediction model. Which is pretty deep subject and I’m not qualified enough to explain it properly (I have computer science background but AI is not my field). About the reliable data part, that’s what the true goal of the AI is to reach the same conclusion as if they had reliable data. Like the pregnant example, the reliable data is pregnancy test, but advertising AI can reach the same conclusion (with certain success ratio) from other unreliable data without the need of the reliable data.

          Simply put, what AWS trying to do is not some random clueless jargons and it makes sense and already used for other things today.

          As for what the % means, yes it doesn’t explain much with only that graphics, but that’s different discussion.

  9. Haha yep they literally just make up stuff to try to produce drama. I actually laughed out loud watching in Brazil that Max had only 30% grip and Lewis chasing had 80% (or thereabouts) and then Lewis pitted. Utterly ridiculous and adds nothing.

    1. I might need to start taking more notice of it. Hadn’t realised the comedy value :D.

  10. Lenny (@leonardodicappucino)
    25th November 2019, 10:09

    I feel like the biggest issue with the graphic is the meaninglessness of the percentages. Like obviously 100% is brand new, but what is 0%? Is that the point where you can start getting punctures from wearing out the carcass? When the tyre is x seconds slower than brand new? Like what is it. I like that they’re trying things with new graphics, but without knowing what the percentages mean, the graphics are meaningless. It’s like saying it 75 degrees outside, without saying what temperature scale you’re talking about. It could be insanely cold (Kelvin), very nice weather (Fahrenheit), or burning hot (Celsius).

    1. Love the way you show how this statistic is pretty useless currently with the temperatures @leonardodicappucino :-)

    2. @leonardodicappucino Don’t give them ideas to put all available units! Damn they are already doing it with speeds, and in a recent bid to attract boat owners, they will now add knots as well.

      On the serious note, they mentioned it is a performance index based on publicly available data. Could it be as “simple” as computed based on lap time, current vs ultimate? If they took Ver and Ham quali lap plus time for the extra fuel at start of GP, then obviously if they lap at the same pace during the GP, Ver will show much lower tire perf than Ham.

      Disclaimer, I am not part of the graphic team and my computation might be more accurate than actually shown on screen.

      Actually I found the thermal imaging of the tire we had a few years back much more interesting than those %.

  11. Just get shut.

  12. I think any graphic that requires a trip to Google to get a vague idea of what it’s trying to say, and why it’s saying it, is a bad idea.

  13. Crofty loves the tyre-graphics, Brundle does a facepalm every time it shows up on screen.

    1. I’m normally the first one to rip into Crofty’s commentary, but i’m almost sure he’s acting like that because he was instructed to.
      I mean, have you noticed he never neglects to mention AWS?

  14. Some of these new graphics are bloody awful and badly thought out. the tire one is useless and the ‘predicted’ gap is always wrong. the ‘striking distance’ graphic is entirely unnecessary.

  15. Most of these graphics will be unnecessary if/when the teams are hopefully more competitive in 2021.

  16. Typo in the titllle…

  17. Completely unnecessary graphic. I really liked it a few years ago when you could see which tyre, new/used and number of laps (tbf just which tyre and how many laps it’s done would be enough). In a data driven sport, the people at the top seem to forget that a little mystery and suspense provide a significant proportion of the entertainment

  18. I think pirelli is confused and totally out of touch want viewers actually want. They held on to the complete failure of 9 or 10 tire compounds despite the majority not wanting that nonsense. Now they don’t want people to know that tires wear during a race. People get confused about car going slower because their tires are more worn? Just stop talking. Just zip it. tssshh! The less fia and fom listen to these folks the better.

  19. The graphic is not confusing, its pretty easy to understand. Its whether the data is accurate is the issue.

    Thats said, if they can get it with accurate data then I love the idea the graphic itself is pretty cool and I love charts and graphics in F1 (that’s probably the video famer in me though).

  20. It’s an oversimplification (and often just wrong) rather than difficult in any way. Just give us core- and surface temperatures and tyre-age. People understand that while they are busy driving, even in casual, non-core-audience arcade-titles like the codemasters-game. There’s no reason to believe they can’t understand it while not actively having other tasks at hand. It’s really odd how TV so constantly and conspicuously underestimates viewers capability of digesting data.

  21. No, I don’t find it hard to understand.

    I totally ignore it. Like all fake data.

  22. I’m finding it really difficult to care anymore. Just hurry up and finish ruining the sport so I can move on.

  23. Disregarding how I feel about the graphic I think he has a point. It’s not tire wear, and I think people confuse it with tire wear, that every lap it goes down a bit, until you have to pit. In reality it’s more like a rapid decent from 100% for the first laps and then it settles in period where if falls a bit or even stays the same, like he mentions. You can have tire degredation and stil have the same performance. Though they should avoid saying performance life, cause that is not helping.

  24. If the figures are lap-time-related, then they are easily manipulated by drivers pushing/not pushing. Drivers tend to push hard right before boxing, tricking the AI into believing the tires are still in excellent shape, while in reality they are near the end of their useful life. So the graphics are most likely meaningless.

  25. I have an opinion
    26th November 2019, 9:55

    Mario, the fans here understand the graphic better than you. We know it’s plain rubbish.

Comments are closed.