I have a computer model, which I use for predicting the weather. The algorithm is fairly straightforward and goes something like this:
It will rain tomorrow.
(Round where I live that's likely to be a pretty effective prediction.)
Anyway, if I run my computer model repeatedly, I find that 100% of the runs give the same result - "It will rain tomorrow". I conclude, therefore, that we can say with 100% confidence that it will rain tomorrow.
I thought of my little computer model when reading the 2014 progress report of the Adaptation Subcommittee of the Committee on Climate Change, which describes the UK's progress in managing climate risks. In it, we learn that:
It is very likely (90% confidence) that greenhouse gas emissions over the 20th century substantially increased the risk of flooding in England and Wales in autumn 2000.
The citation is to Pall et al, a paper by Myles Allen's group, the key finding of which appears as a footnote in the ASC report:
The precise magnitude of the anthropogenic contribution remains uncertain, but in nine out of ten simulations 20th century anthropogenic greenhouse gas emissions increased the risk of floods occurring in England and Wales in autumn 2000 by more than 20%, and in two out of three cases by more than 90%.
So you see, while you may have sniggered at my little computer model, my approach is legitimised by none other than Lords Deben and Krebs and Professor Myles Allen.