Dawkins' scale and the spectrum of beliefs about God

I have some problems with the often-cited Dawkins scale for describing the spectrum of beliefs. The scale is described here: http://en.wikipedia.org/wiki/Spectrum_of_theistic_probability#Dawkins.27s_formulation

I think it represents the probability distribution assigned to the two propositions "god" and "no god", but fails to represent certainty attached to those distributions. In practice I think almost nobody would say, "I believe there is 50% chance god exists and 50% chance god doesn't exist, and I'm 100% certain these probabilities are correct".

So, I propose an alternate representation for belief distribution:

* technically atheistic positions are restricted to the left-hand line of the triangle, but people may self-identify a little differently in practice.