Saturday, October 29, 2011

F1 2011 India Qualifying Report

My stats graphics qualifying report for the 2011 Formula One Indian Grand Prix qualifying session can be found at: F1 2011 India - Qualifying Stats Graphics Report


  1. I don't know I shall be happy or sad on entry of India in elite's club "F1 club".On one hand I feel proud of being an Indian at this moment, on the other hand I see some paradoxes produced by some latest data (1) on crimes which says that in 2010, 5000( -) children were kidnapped, raped and killed (2) Quarter a million farmers have committed suicide in a decade; 17 infants died in a state hospital in just 48 hrs. Millions of people sleep daily empty stomach, millions of homes don't have drinking water and proper sanitation. Millions of kids are still deprived of schools. Besides, all these and many more paradoxes, we say that India is shining, we have f1 track, organised commonwealth(guess successfully), going to become superpower, aspiring for hosting Olympics blah blah blah...... .Really, but on what cost? We feel proud of entering any elite club but forget people behind us. Is it law of nature or man made paradoxes?

  2. I am totally agree with you rahul that india has done great job to be part of f1 club but the same time worry about this conditions.

  3. This was nice but India first grand prix is great and was fun to watch this all series.

  4. Hi Tony,

    I like the automated reports. I'd like to think that as a race progresses, we can analyse the data and update predictions based on some assumptions, some data fitting (intelligentF1 model) and some optimisation routines (prediction of pitstops based on race state - I already do this, but it's independent of the model).

    I'm moving slowly towards a coded version (js/flash) and I now have a basic traffic model. I can dynamically replay the race and watch it (and my predictions) evolve. But I can't post flash on wordpress, so it's not 'out there' yet.

    If we can get this to output key race information as it happens to inform the viewer (like a predicted loss/gain when Webber switched to three stops) then this has the potential to be a serious bit of datatainment.

    We can get the data as it happens now, but there are key bits missing in the live timing feed, so there's a bit of work to do. It's fiddly, but it can be done.

    Do you think there's some way to work together on this?


  5. Hi James -

    Been loving your intelligentF1 posts and regretting not finding an hour or two to work up some new charts eg based around suggesting primitive models for each stint, trying to pull out in/out lap times on pit stops etc (my working knowledge of R is minimal, which doesn't help;-)

    Trying to work on a live datatainment view sounds really interesting... one thing I keep meaning to do is try and generate some views over the data that I scrape into a scraperwiki database once the FIA timing sheets are released (of course, this isn't live, and takes maybe 40 mins after the end of a session for the FIA to post PDFs; then time to fix broken scrapers;-) This would be okay for post race browsing... (I also clear the scraper just before the start of the next race, in part to respect the FIA license conditions..)

    Being able to view possible strategy changes based on 'events' during the race could be really interesting, allowing the fans to maybe come up with better predictions than the commentary teams, or at least, get them playing along as armchair strategists. With a little bit of crowd sourcing, and maybe twitter analysis, we may even be able to give best guess on current tyres, as well as having access to summaries of pit events to date from timing data?

    We're in the middle of a house move atm, but as soon as life returns to normal, I'll try to pop up some examples using Javascript libraries (which similarly won't work in blogs;-) I've found blogger, for all it's other sins, does let you stick pretty much anything in a page.)

    Btw, have you had a look at the McLaren telemetry at all? A friend has been collecting most it for a couple of seasons now, so if there's anything you think might be useful in there?

  6. @james in passing, do your flash swf files convert to HTML5 using Swiffy?

  7. Tony,

    In my ambitious moments, I'd like to think that the live datatainment tool could be a support to commentators - not sure how much interest there would be, but at Webber's third stop, I could show a little graphic showing I think he loses a couple of seconds theoretically but doesn't have to pass Massa on track, giving an opportunity for commentators to, well, commentate.

    I'm pretty confident that I can tell which tyres a car is on from the model as long as the fuel model is OK and I know which tyre they started on - I'm working on getting the fuel load refined automatically based on the first few stops seen on TV (so we know which tyres they go on). I can code the model to do the rest (I think). Unless the compounds get really close in pace next year...

    Never played with the McLaren data - there may be some good clues in there. Depends on signal-to-noise on anything interesting.

    I'll try Swiffy. But it's all got a JS back-end.

    And perhaps it's time I tried the twitter thing.



There seem to be a few issues with posting comments. I think you need to preview your comment before you can submit it... Any problems, send me a message on twitter: @psychemedia