Visualizing T.T. Finals at NID Bangalore

Visualizing T.T. Finals at NID Bangalore

November 8, 2012 12:01 AM 4 comments

Last month, the (newly elected) sports committee of NID Bangalore organized a Table Tennis Tournament. Now I’ve started playing table tennis quite often at NID, so I happily participated. Defeated in the quarter-finals by Amal, I was watching others in awe until I got a random idea. Visualization time!

So here’s my first attempt at making a data visualization: Final matches between Amal & Upamanyu. Click to enlarge!

TT Finals DataViz by Rasagy

Final Visualization (Click to enlarge)

Still here? Awesome. Here’s a quick look at how I made it:

The idea was impromptu – I had not planned it, so I didn’t think much about how to record data. I decided to simply make dots: One for each point, in two columns. I kept recording scores after 5 serves (when serve changes) to tally my data.

TT DataViz - Recorded scores by Rasagy

Scores recorded during the Final matches

 

This easily translated into a string of 0′s & 1′s: 0 for Upu’s point, 1 for Amal’s point. I typed these into Excel, and let the conditional formatting do some magic.

TT Data Viz - Excel Data

Conditional Formatting FTW!

Sweet, eh? The columns are individual points (3 columns), cumulative score of Amal & Upamanyu (4 & 5), relative score (for each of the three matches) It’s the relative score that seemed quite interesting to me (It also highlighted the difference in each match instantly!).

For more ideas, I tried a few charts. I wanted to highlight two things: the continuous streaks of points, and the relative difference in score throughout the game.

TT Data Viz - Excel Charts

Playing with Excel Charts

My favorite? A simple line chart (Others seemed harder to understand/biased to one person). Technically, the correct chart to use should be a bar chart (since points are discrete), but the chart looked horrible in Excel (I tried sketching a mix of line & bar chart, which looked more like stairs than a graph). Maybe I’ll try a more honest chart next time!

With this decided, I tried to re-create the line chart in Processing. The advantage was that I could plot it serve by serve, maybe turn it into an interactive visualization later? I also decided to add the relative scores along the X Axis (To emphasize on the lead a bit more). Circles seemed to stand out a bit, which was just what I wanted. Here’s a quick gif animation of the first match.

Finally, I exported (saveFrame) from Processing, added some text in Photoshop, and voila! Here it is!

This is my first attempt at data visualization (and at making something useful in Processing!), so I’m sure there’s a lot for me to learn. Was the chart hard to understand? Or is there something I could’ve done better? Feel free to comment! :)

PS. Special thanks to these people:

Here’s hoping this is the first of many visualizations. *fingers crossed*

  • http://profiles.google.com/ccluri Chaitanya C H

    Hey Rash, 

    You can also visualize these
    1) A table in the bg of the chart that indicates points won in forehand versus backhand of each player.
    2) Points won my self versus points won because of opponents mistakes (opponents nets the ball / equivalent to self goal in football)
    3) Points won in offense versus defense. 

    I didn’t know you were interested in data visualizations. I have some cool ideas, let me know if you are interested.

    • https://twitter.com/rasagy Rasagy Sharma

      Chaitanya,
      Nice ideas. Next time, I’ll try and record more data, so such insights can be gathered. Also, how do you label a point won in offense vs defense? (Maybe there’s a technical meaning I don’t know of).

      Anyway, I’ve recently got interested in data visualizations. If you have any interesting idea, feel free to let me know. :)

  • Mainak

    Hey Rasagya!

    Great to see this post on your blog. Check out this blog: http://junkcharts.typepad.com/junk_charts/ if you’re more interested in data visualization. Your graph looks beautiful obviously. Keeping track of the stronger player on the x-axis is a nice idea too (contrast could be higher maybe?).

    Another idea that occurred to me was to me was to check out the trend across matches for the two players separately. Oh, and next time – do remember to note who served (do you have that already?). It’s happened to me a lot when I play TT that I get all points while serving but not the other way round. That could be interesting too!

    • https://twitter.com/rasagy Rasagy Sharma

       Finally found your comment. :D

      Junkcharts blog looks great, will follow it regularly from now on. :)

      I do have the data of who served when (5 serves each, so wasn’t hard to note down). However, I didn’t see any insight (Hardly any point was earned in the first shot). Plus, since I didn’t really have much data on which shot *after* the serve the point was taken, it didn’t seem very insightful.

      Next time, I’ll try and collect more data. Let’s see how that goes!