Visualization is King

featured, Performance Analysis — By on January 4, 2012 12:44 pm

Over the last 6 months I have been working on quite a big data collection project.  I have been doing my best to find innovative and interesting ways to present all this information. I am still convinced that the role of an analyst is to get a message across as much as it is to collect data/videos.

I recently came across a couple of blogs I hadn’t seen before – both deal directly with Data Visualization and I wanted to share a really interesting example of what I found.

A Table of Results

Both blogs were dealing with the Manchester Derby from a few weeks back where City won 6-1. On an excellent blog soccerbythenumbers a table of the historical results were presented like so;

Nothing especially wrong with the table but nothing very exciting about it either. I would imagine most people take a quick glance at this and then move on.

Simple but effective improvements.

On the VizWiz blog they looked at making this table come a bit more alive and could use the same data to tell their story in a better way.

So how do you think they did? Here is what they did.

  1. Changed the numbers to percentages and rounded to one decimal.
  2. Removed most of the grid-lines.
  3. Formatted the results as a heat map.  This makes the largest percentage of result very obvious.  For example, you can now easily see, without having to scan across all of the data points, that 1-1 is the most common score line…boring result!
  4. Formatted the totals as a second heat map.  I chose a brown-white scheme for these.  The totals show you the % of the total goals scored for each time.  ManU has scored one or two goals 64.2% of the time while City has scored one or two goals 58.2% of the time.

For me this paints a much clearer picture of exactly the same data. But that’s not where the improvement ends.

Visualization is King!

On the dataremixed blog I found an even better improvement. Using a bubble chart they managed to create the following interactive and visually appealing table – again using the exact same data set. (To see the table in all it’s glory click here)

Think like a Storyteller.

This takes the visualization of the data to a whole new level. You now have colours representing the 2 teams and a draw and the size of the bubble shows how popular the scoreline is. I think any human being would prefer looking at the last table than the first. But more importantly that what my preference is – I am also much more likely to remember the important points.

I am constantly on the look out for new and exciting ways to present information. I think the example above highlights just how important that skill is.

Tags: , , , , ,

12 Comments

  1. Ash coates says:

    As with everthing a picture paints a thousand words. We must remember as Analysts half of our battle is engaging our audience with presentable stats which grab there interest. This article shows some excellent examples of displaying our product

  2. Neil Melville says:

    Another suggestion for the second table (the one with heat mapping) would be to hide all zero values. You can already see how well this works in the bubble chart. In Excel this is done in the Advanced options:
    Application button > Excel options > Advanced

    Because this choice applies to the entire worksheet, and that may not be desirable for other tables, another alternative is to create a custom number format that displays a dash instead of a zero. e.g:
    0.0%;[red]-0.0%;”-”

  3. Andy Kriebel says:

    Rob,

    Thanks for the mention. I agree with Neil that my heat map would benefit from hiding the zeroes. Unfortunately, I didn’t save the Excel file or I would make that change.

    Since you’re seeking out nice presentations of soccer data, you might find another recent blog post of mine interesting.

    It analyzes what it takes to survive in the EPL and is interactive. I think you’ll like it.

    http://vizwiz.blogspot.com/2011/12/what-does-it-take-to-survive-in-english.html

    Andy

    • thevideoanalyst says:

      Hi Andy – I have seen that interactive table before – its fantastic. Anybody who hasn’t seen it should check it out.

  4. Nigel Donohue says:

    Is there any statistical/visualisation/dashboard software on the market that can be used to input and present data as you need it or is Excel still the best option?

    • thevideoanalyst says:

      Nigel I still use Excel but would like something a bit better than that. The software used in the above example is Tableau – they have a trail version on their site.

  5. Chris says:

    Graphs & tables are ok for analysts and sports science personel but I think that bullet points summarising key points are more what coaches are looking for.

    • thevideoanalyst says:

      Do you really think so Chris? I would have thought a certain times making the data more visual is of benefit to everyone.

      Although you should know your audience I guess.

  6. Ben Jones says:

    Hi Rob – thanks for picking up on my project and commenting on it! Tableau Public is a free tool anybody can use to create interactive visualizations like the ones Andy Kriebel and I make. There’s a great online community and I’m just getting started really. I’ve learned a lot from people like Andy – not just about how to use the software, but about how to make data come alive. Thanks again, Ben

  7. Laban Scott Libby says:

    Hi Rob,
    Nice collation of the three takes on the same data – shows how easily you can utilise it more effectively.

    I’m sure you will have seen it already, as it is from VizWiz and Chris reposted it on SBTN, but the visualisation of the PL table from 01/02 onwards is another great example of taking simple data and making it much more interesting and useful, I think. Especially as it uses a free online visualisation tool.

    http://vizwiz.blogspot.com/2011/12/what-does-it-take-to-survive-in-english.html

    Cheers, Laban

  8. Keith Lyons says:

    Rob

    Thanks for the post and alert to examples of good visualisation. Edward Tufte’s work fascinated me and I still have a copy of The Visual display of Quantitative Information from the 1980s.

    I have been looking at the opportunities to analyse and visualise secondary data.

    I wrote this last year to link to s US Post service http://keithlyons.me/2011/05/09/medal-predictions-2012-olympic-games/ I like the idea of a simple enquiry producing detailed information.

    I note that Big Data are an IBM priority and I have been interested in their Many Eyes service http://www-958.ibm.com/software/data/cognos/manyeyes/

    Thanks again for the post.

    Happy New Year

    Keith

Leave a Comment