Now With Added Pie

About a month ago I posted about Ed Reardon's Week and how long it had been since series four had been broadcast. Radio 4 then started broadcasting series eight a week later and I felt a bit stupid. Still, I'd tried out a few graphs before making that post and I ended up going with mostly bar charts because I thought it was the best, or probably only, way to represent the data.

Well, my anonymous commenter requested a pie chart - something I deride on an almost daily basis at work - and it made me think about whether I actually could use a pie chart to represent the number of broadcasts without losing the time component of the dataset. This was the result.

The chart (and all those that I post below) come with a massive set of caveats: The graphs represent distinct broadcasts of a series based on data from the relevant BBC page, though I've discovered that there are definite inaccuracies on those pages. Broadcast dates are taken from the last episode in a series. There is no data from before the mid-2007. The overall sector still represents the number of distinct broadcasts but the coloured areas have no direct significance. Each radial band represents a six month period.

Put simply: this isn't scientific. Think of it as a simplistic infographic that gives visual weight to more recent broadcasts.

Anyway, I realised almost as soon as I uploaded it that the chart was still inaccurate because it doesn't account for the original broadcast date and so I've now altered it slightly to remove the sectors which represent time before the first broadcast. Once I'd done this for Ed Reardon's Week I knew I'd have to do the same for more programmes because... Well, mostly because I'm not actually capable of leaving things like this alone.

The charts are  way more effective where there are a good number of series but I couldn't leave out Cabin Pressure and Bleak Expectations. And, just to make it clear, I'm not making a point with any of these (except for Ed Reardon  - series four please!). I'm just a curious data geek.

Bleak Expectations

Giles Wemmbley-Hogg

The Maltby Collection

Old Harry's Game

Cabin Pressure

Ed Reardon's Week

Oh, and I'm sorry for the inconsistent labelling and font sizes. I don't have an excuse except that I did the charts over a number of nights, on some of which I got bored of labelling.

A big thank you to my anonymous commenter for making me challenge my natural contempt for pie charts!

Ed Reardon's Week Statistics

I've been listening to some old episodes of Ed Reardon's Week over the last few days. I'd forgotten how enjoyable it is, even though (or maybe because) half the references go clean over my head. 

For a while I was considering an attempt at a Reardonesque diatribe for this post, complete with disparaging references to Radio 4 Extra (extra what, precisely?) and its schedulers (no doubt composed entirely of twelve year-olds) but neither my vocabulary nor my vitriol is up to scratch.

And so I shall resort to my own medium to make my point: data.

There have been seven series' of Ed Reardon's Week, running between 2005 to 2011. I've looked at the BBC programme pages and the number of broadcasts each series has had. The initial broadcasts of series one to three pre-date the records so I freely admit that this isn't exactly a rigorous, "More Or Less" level analysis.

This graph shows the number of times each series of Ed Reardon's Week has been broadcast since 2007 (1-3) or since its first broadcast (4+). I've eliminated direct repeats but included repeats on 4 Extra that followed more than a month after an initial broadcast on Radio 4.

Well, that doesn't look terrible but it's not exactly comparing like with like. Series seven has had the same number of broadcasts in 14 months as series four has since November 2007. How could I express that as a graph, you might ask. You might, but no reasonable person would.

Nevertheless the answer is, like this:

Due to the fact that I'm still not quite up to speed on Excel 2010, the calculations aren't as elegant as I'd have liked but this graph is attempting to show a sort of broadcast-per-day for each series. 

The time period for each is measured from the date of the first broadcast recorded on the programme page (so the caveat stands about series 1-3).

I realise that this post has turned into a exercise in how to thoroughly alienate your reader but the point I'm trying to make is this: 4 Extra, please stop repeatedly broadcasting the first two series' of programmes. Ed Reardon is only one example of many great comedies that are stuck in a repeating loop of their early years. Occasionally the third season comes along after a six months or so (Vent!) but, more often than not, series one starts again a year later.

It's okay if this graph makes you feel that I need to get out more.
It makes me feel the same way.

I do realise that 4 Extra is all about repeats. I love that! My listening experience would be a lot narrower without the fabulous range and depth of the 4 Extra schedule. I also realise that new listeners are probably more likely to stick with something if they can hear it from the beginning

But please, I'm begging you 4 Extra, consider the graphs, look at your own records. Series four hasn't been broadcast for four years. Surely longer-term listeners deserve the reward of something new from time to time? 

Because the alternative is more graphs. And no one wants that.