Sunday, 11 December 2011

Regression to the Mean

In which I almost entirely mis-apply a statistical concept to make a point about writing. I'll point out right now that although I'm quite interested in statistics, much of my knowledge of them was obtained through the medium of cricket. My knowledge of regression to the mean, specifically, through an article by Gideon Haigh on freakish early career averages. The notion, as I understand it, is that over time the average of an expanding set of data is likely to become closer to any general average. Things become more averagely average as they go on, if you like. Also, anyone touted as the new Don Bradman will probably suffer a string of ducks to bring them back to the realms of mere mortals.

Why does this have any relevance to writing? Because of series, of course. Think of all those long series out there. Think how wonderful the first books were. How exciting. How different. Now think of the later books. They might not be bad, exactly, but they do tend to be far less original. They come back to the pack in so many cases. Why is that? Well, the first reason is that often the author isn't writing something they have been incredibly inspired to write, with all the unique plot elements that come from that. They're working more from the craft of writing. There's nothing wrong with that. Not at all. But it means they're less likely to come out with something new.

So those of us with series in our heads should probably think long and hard about them before writing the next book. I'm not saying avoid it. I'm just suggesting that we should only ever write that sequel if there is genuinely something we want to say with it that sets it apart. Otherwise, we're just drifting back, and I'm sure your writing deserves so much more than that.

No comments: