Thanks to Mike Pickard for correctly identifying that this is the sort of thing that I would be interested in and pointing me in the direction of twitalyzer.

Can you hear me say Yay! Pointless statistics about twitter!

Twitalyzer breaks down your Twitter feed into five categories: Influence, Signal, Generosity, Velocity and Clout.

Influence is the ‘overall’ metric, built up from combinations of the other ones. As such, there’s not much explanation needed for this one. It is perhaps not surprising to discover that I have less influence than @StephenFry on twitter, since I have 66 followers (although I could be up to about 80 by now if I’d not blocked the spam accounts — tip: if you call yourself “Holiday Offers”, I’m likely to assume I won’t be interested in what you have to say), whereas our Stephen has, at the time of writing, 311,598 followers.

At first glance, this would appear to indicate that he has a lot more twitter followers than @ThePickards, but I’d rather look at it from the point of view that if we assume the world population is 6.5 billion, he is only reaching 0.0048% of the population more than me. Just a drop in the ocean, isn’t it? We’re virtually the same…

And the second statistic ‘signal’, shows why there’s so little difference between us. This is the measure of signal to noise. The algorithm, while presumably not perfect, is there to try and establish that people are passing on useful information which may be helpful to others, instead of just people whinging about train delays.

Sure, Stephen has been on telly a bit, but it would appear that according to these statistics, that what I have to say is more worthwhile. Just look at that. Nearly 90% signal to 10% noise for me; 60-40 for the Frymeister. Stephen Fry, all of your base are belong to us.

However, the next statistic, Generosity, shows neither myself nor Stephen in a good light. Generosity is rather simplistically measured as the amount of time you retweet someone else. I think that’s a little simplistic, to be honest — if I were to retweet something Stephen said, it would be a bit pointless, because I’d be broadcasting it to a significantly smaller audience. Where Stephen does it, he broadcasts to a significantly larger audience.

There is therefore an appreciable benefit in someone with a lot of followers retweeting, and much less point in someone with only a handful of followers retweeting. This statistic therefore needs mucking about with, I would suggest, to take account of the ‘reach’ of the person re-tweeting, and also to assume that you don’t actually want every single tweet someone produces to be a retweet…

But, according to this statistic, Stephen and I are remarkably un-generous, with relative generosities of 1.1% and 2.6% respectively. We are stingy twitterers.

So far, Stephen is scoring far more highly than me, although I’m actually beating him in the first two categories. How can this be?

Oh wait, perhaps he’s ahead of me in the next two categories…

Next up is velocity, which is simply a measure of the number of tweets you produce, against a measurable maximum of 1,500 per week. So if you are producing more than 200 tweets per day, you’ll be doing fine in this regard. Only you might be decreasing your signal-to-noise ratio.

Unless of course you post 215 retweets per day, which will give you the maximum ‘generosity’, maximum ‘velocity’ and maximum ‘signal’. As I understand it, this should be sufficient to ‘game’ the twitalyzer and score you top marks in each of these categories. The minor disadvantage is of course that you would piss off anyone who was actually following you…

And of course, Señor Fry is ahead of me here, with a velocity of 25.3% to my 10.3% (equating to some 54 tweets per day, if I’m reading that correctly for Stephen, as opposed to a much more feeble 22 tweets per day for me). Of course, I would like to point out at this juncture that my tweets are more worth reading, as over 40% of Stephen’s tweets are apparently just ‘noise’.

Somehow I doubt that I’ll convince his 311,000 fans of this, though…

Clout is simply the amount of times you are referenced; how likely it is that someone else will reference you in a tweet. Not surprisingly, Stephen is ahead of me with this one, with a clout which theoretically could not be bettered. On the bright side, I’m only 98.1% behind him, and it’s unlikely that the gap can get much wider.

On the down side of course, there’s a fair bit of catching up to do.

But I’ve had fun playing with the twitalyzer, even if I had to do the side-by-side comparisons myself: if I could specify one feature request, it would be to allow side-by-side comparisons. That’s where the real fun is, not just in looking at some bland statistics but in having a ‘competition’ with someone else, regardless of whether you take that competition seriously or as a joke.

I’ll let you draw your own conclusions about how I was treating it…