Tiger Woods & Why BI Tools are Awful

Recent readers of this site may not realize that last year I wrote an OpEd for the Wall Street Journal on the trouble with Tiger Woods: I argued he swings too hard. Without getting over-deeply into the data — other than to say that I’m an expert at swinging too hard in golf — I thought it would be fun to show the change in Tiger’s game (and in his competitors) graphically.
Check the following animation (in Flash format, so refresh to see it again) I put together — Tiger is the red dot:


As you can see, despite the supposed improvement in his game, Tiger’s driving accuracy has fallen off much more than his driving distance has increased. At the same time, the gap between Woods and the mass of his professional competitors is much smaller than it was five years ago — the animation shows this nicely, with a great splotch of competitors all closing in on Woods over the period.

As an aside, this is a good example of why current BI and analytical tools are woeful. Why can I not do this sort of dynamic analysis in Excel? Why is Excel stuck analyzing dead, static data? While this wasn’t difficult to do by pouring some data and images into Flash, it was more trouble than most people would ever go to, despite the obvious insights gained from seeing the data change dynamically.

And why do I have to use Flash at all? Why not, say, Ajax? Then again, while Flickr is apparently making the jump from Flash to Ajax, it strikes me as early and unnecessarily brave. After all, at least there are IDEs (sort of) for Flash with which you can build UI-centric apps; the same is affirmedly not true in Ajax. We have a long way to go before we get Ajax-oriented BI apps, no matter how good an idea that might be …

Related posts:

  1. Tiger Woods & Echoes of the Dot-Com Boom
  2. Tiger Woods, Candy Stripers, and Fire Survivors
  3. Me, WSJ, and Tiger Woods
  4. Tiger (and Nike) are Free Riders
  5. Separation of Church (Tools) & State (Storage)

Comments

  1. We got IMAX, ExLax, and now Ajax. WTF is next?

    We got IMAX, ExLax, and now Ajax. WTF is next?"Where can you find Flickr and Apple in the same room? At the AJAX Summit of course. An

  2. Tracking Tiger

    The very bright Dr. Paul Kedrosky has posted an interesting look at use of BI tools. Tracking Tiger Woods drive accuracy. He makes a great point of the lack of evolution amoung XL and other tools.
    Just think what this could do for motivating…

  3. For those wondering, BI stands for Business Intelligence.

  4. Weekend Notes

    Dave Pinto was at RFK Stadium on Saturday night, and he has video. And here’s hoping they throw the book…

  5. Drew says:

    This visualization would be radically more useful if Flash wasn’t doing the morph routine on the graphs and screwing them all up. Really the only thing I can make out from this is what Tiger’s trajectory has been over the last 5 years. That’s valuable, but you don’t need animation to tell that story. This would do much better as a static small multiple (which are annoying to make in excel, but possible). There are very very few situations in which animation is the right visualiation tool to use. This is definitely not one of them. To alleviate this, it would be fantastic to have a slider that lets you pick which timestep you want. The jarring point at the end of the cycle makes it really hard to understand what’s going on.
    It seems like there is a tradeoff between distance and accuracy. What strikes me as particularly interesting is that Tiger is trading more accuracy for less distance than other players – it just happens that he had more accuracy to begin with. That seems like a much more valuable story than the morphing blobs of points we have now.
    I appreciate the attempt, though. It’s definitely interesting data with some cool stories to tell. This just doens’t seem to be the best way to tell them, though.

  6. Lucian Beebe says:

    Surprised by your comment that animation is rarely helpful in data visualization. The problem here isn’t animation. That’s effectively used. Its interactivity. With a bit more effort, I could speed and slow the rate of change here, pause and restart, hover over the competitors to see who they are, turn on and off trails, zoom in and out, etc. Then, we’d have ourselves a great BI visualization. All that can and should be done in Flash to make it great.
    My favorite in this category is still the somewhat old World Health Chart: http://www.gapminder.com. Great data visualization with complete user control. Its done in Director, though could just as easily have been done in Flash or Flex.
    Full disclosure: I do work for Macromedia :)

  7. Robert Schwartz says:

    And why do I have to use Flash at all? Why not, say, Ajax? Then again, while Flickr is apparently making the jump from Flash to Ajax, it strikes me as early and unnecessarily brave. After all, at least there are IDEs (sort of) for Flash with which you can build UI-centric apps; the same is affirmedly not true in Ajax. We have a long way to go before we get Ajax-oriented BI apps, no matter how good an idea that might be …
    Sure. Whatever you say, Boss.
    BTW what language was that?

  8. John Davison says:

    Hold on a moment there! To me the animation suggests something about the field as well as Woods. Notice that the others are comparatively tightly clustered in both distance and accuracy at the outset, but the cluster explodes over time. My eyeball suggests that the centroid of the cluster exhibits the same tendency toward decreasing accuracy with increasing distance but it seems clear that some have improved in both regards. Here is where interaction could really help by enabling one to pause the animation and query for quantification as well as individual details.

  9. John Davison says:

    Also, vis-a-vis accuracy vs. distance, a small angular error in deflection will result in a larger absolute linear error at greater range, so it might be useful to be able to assess both types of error. Also, does accuracy measure error in both range and deflection? Perfectly straight is not very useful if range error drops the ball in a bunker or water hazard.

  10. biff says:

    You might take a look at Spotfire DecisionSite ( http://www.spotfire.com ). It costs around $2k or so, but it puts an awful lot of usable power in the grasp of mere mortals. The Tiger Woods example would be trivial to do via Spotfire.

  11. Jasmine says:

    I am currently working on tools to do exactly that, better analysis without having to know programming languages… My question is… where did you get the raw data for that? I usually do my own data entry! I’m typically only looking at my own game so it doesn’t matter much, but I would love to get a big set of raw data from the PGA… where did you get this?

  12. k says:

    Data can be found under Stats at PGATour.com