More Lying with Numbers

Not willing to leave well enough alone, I decided to go back to the StatCounter browser statistics and spread the time limit out a bit. Like three years. And here's what I discovered buried in the numbers from August 2008 to September 2011.

Source: StatCounter Global Stats - Browser Market Share

You don't need a fancy chart to understand the numbers on this graph. From August 2009 to September 2011 (now), Internet Explorer has collapsed in market share by nearly half, from 58% in August 2008 to 33% now. Firefox has essentially remained flat at 35%. The big story is how Chrome rose from nothing to 22%. The rise of Chrome has had a profound influence in Europe, drastically reducing Internet Explorer (and to a smaller extent Opera) while essentially keeping Firefox flat at around 35%.

But what's interesting is the far right where it looks like Internet Explorer and Firefox crossed again. Let's go back and look at the chart, using the same start time as our brilliant scholar over on TM Repository, May 2011, but extending it September.

Source: StatCounter Global Stats - Browser Market Share

Oh my goodness! Would you look at that? Just one month out from July Firefox has caught back up with Internet Explorer. And Chrome has risen a full percentage from July. Internet Explorer is back to dropping again. Why? Who knows? Who cares? It just is.

The fine folks over on TM Repository should stick to creating their postings out of whole cloth. Because when they try to twist real facts, those real facts will always unwind and knock them back on their collective ass.

Update 17 September

When I wrote this I took the snapshot the first of September but left it live. I could have grabbed a static JPEG, as the sight allows for this. But I left it live so that as the statistics developed during the month. As of mid-September IE has actually climbed back up a bit, so that it's 2% higher than Firefox (35% vs 33% respectively), while Chrome has remained essentially flat at a little less than 22%. The world-wide statistics over this same period are a little more close to ground truth, with a three-way split essentially between IE, Firefox and Chrome; 50%/25%/25%. And I think that's the way it will be for the next six months. After that, who knows?

Oh. And this is all desktop. Mobile browser statistics are a lot more varied.


  1. I use both FF and Chrome. I do this because some sites work better on one than the other.

    IE is really only used for me to download a browser on a fresh machine.

    Your graphs are FANTASTIC

  2. Your graphs are FANTASTIC

    Thanks, but that comes from simply copying and embedding a link provided by StatCounter. It's StatCounter that's doing all the heavy lifting, producing the graphs.

  3. There you have it - you can prove anything you will with numbers. Being in the numbers game myself (working for IBM Cognos), I should know what I'm speaking of.

    Oh, our bosses love those numbers. With just the same basic set of data they can:

    a) prove to themselves how good they are, and
    b) prove to us how bad we are, and that we have to work much harder

    It's a farce, and those beancounters are constantly lying to themselves.


Post a Comment

All comments are checked. Comment SPAM will be blocked and deleted.

Popular Posts