Monckton's Fallacy in Five Minutes

As I mentioned in a previous post, Christopher Monckton seems to have contradictory views on when something demonstrates the endpoint fallacy or not. But his Great Pause posts are not by any means the most egregious example of this confusion.

Here is a video of a presentation he gave at Saint Paul, Minnesota. You can watch the whole thing and find many examples of contradictions and nonsense, but the bit I'm interested in starts at around 43:40 and lasts a little under 5 minutes. During those 5 minutes Monckton both condemns one graph as bogus whilst claiming an almost identical graph is genuine.

43:40

Here is the latest lie - in the 2007 report - the iconic lie there. They're attempting to show that global warming is accelerating. Implication that our CO2 emissions are accelerating and this is causing a more rapid rate of warming.

This is a statistical lie known as the start point or endpoint fallacy. Where you take a jiggly up and downy dataset like temperature, where you don't know which way it's going to go next, a stochastic dataset. If you choose your start point and your endpoints carefully enough you can make it look as though any trend you want is happening. Here they've tried to show a rising trend.

I'm now going to take the same data, but I'm going to take the more recent end of it, between 1993 at the present, and I'm going to choose my own start points.

Look at this. Top left 1993 to the present, top right 1997 to the present, bottom left 2001 to the present, bottom right 2005 to the present. We're heading for a new ice age!

In case this isn't clear enough, Monckton now asks his audience:

Now here is a question for you - to see if you've been concentrating as alertly as I'd hoped. Which of those two graphs is nearer to the truth?

Hands up for the UN's graph. None I think.

Hands up for my graph. A few.

Hands up for neither. Much more.

And those who said neither are correct. Because remember I started out by saying this is a bogus statistical technique. Every time you use it, it produces results which could only be right purely by accident. It is invalid whether you do it to try and show a falling temperature or you do it to try to show a rising temperature, and I merely gave you that example to show how easy it is to bend the data in the way the UN has done.

And it is a disgrace that a public authority in receipt of tax-payers money from around the world should dare to produce a graph like that one that we've just seen. That should never have been done. It is a clear continuing instance of deliberate bad faith.

There cannot be any doubt about what Monckton is saying at this point. All the graphs he showed are only to illustrate how bogus the technique is.

That's taken us to 46:00. Now skip forward a couple of minutes.

47:45

So we go on then to a lie which was told by the Director General of the National Climatic Data Center, Tom Karl, when he and I testified along side each other in front of the Energy and Commerce Committee of Congress earlier this year. I produced this graph ...

This graph shows exactly the same period as the bottom left one in his set of bogus graphs. It isn't exactly the same as the one he showed to the Committee, that was this:

In either case the graph shows almost the same time frame as the bottom left one he's already claimed produces results which could only be right purely by accident.

So is he admitting to presenting a bogus graph to the Energy and Commerce Committee of Congress, or is was he trying to make the same point to them, that a graph like this only shows a result that is a pure accident? He goes on ...

... what I'm saying here is that there has been global cooling for the last 8 or 9 years, statistically significant, and rapid cooling. How many of you have seen that reported in any major news medium recently?

We are now up to 48:30, less than 5 minutes since he first described the IPCC graph as a lie. He's admitted to showing a graph to a Committee of Congress that covers the same period as one he claims demonstrates the endpoint fallacy, but is also claiming to his audience that it demonstrates rapid and statistically significant cooling.

You can see the whole hearing here. Monckton's testimony start's at 1:21:30 - though the video doesn't capture the slide in question.

Again he says that the graph shows seven years of cooling, with no suggestion that it is based on a statistical lie.

There's a lot more that could be said about Monckton's recollection of this hearing, but that's moving outside the scope of this post, which was merely to draw attention to Christopher Monckton's inconsistent notion of what makes for a bogus statistical technique.