Why James Lewis is Probably a Bullshit artist

Submitted by naught101 on Sat, 02/03/2007 - 22:42

the post was made in response to a forum post using James Lewis's article "Why Global Warming is Probably a Crock" on american thinker ( http://www.americanthinker.com/2007/01/why_global_warming_is_probably.h… ) to discredit climate science.

--- response part one---
there's a serious flaw in the logic of that article:

see, that only works if you want to be 100% percent correct. the same logic says that with the same variable, they have a 60.5% chance of being right half the time, 47% of being right 3/4 of the time, and 77.7% chance of being right a quarter of the time.

and when I mean right, I mean exactly right. which means that if they are close to right, that would fall outside those probabilities. if they are completely wrong... well. let's have a look at it, shall we?

100% wrong = 1%^100 = 0% (my calculator doesn't go that small. but practically, NO chance of them being 100% wrong)
75% wrong = 1%^75 = 0%
50% wrong = 1%^50 = 0%
25% wrong = 1%^25 = 0.0000000000000000000000001%

obviuosly those figures are complete bullshit. but they use exactly the same logic as the original statement.

one problem is that extremely basic probability maths is not suited to a "hypercomplex system". the other is the concept of "right" and "wrong" in a system of probabilities.

James Lewis either needs to go back to university (or even highschool), or more probably he's a bullshit artist.

-- Part 2 --

sorry. I was going to post a bit of evidence (something the quoted article is extremely short on) for my case.

I assume that this James Lewis character is referring to the IPCC and others in his attack. it's interesting to not that the IPCC's third assesment report actually fell a fair way short of the mark:

as you can see, the probability of them being right (it is happening) had little to do with it, and they actually underestimated (using highly complex computer probability models) the real figures.

Originally posted as a response on OpenDemocracy.net: http://www.opendemocracy.net/forums/thread.jspa?forumID=179&threadID=47…