My Mistake, Corrected

In the ongoing saga with the blogger andtheresphysics (Anders), I’ve come to realize I made a mistake. Naturally, I want to clear that up as soon as possible. So here goes.

The issue of focus came about during an exchange the two of us had in an earlier post here. Anders had said:

What seems indisputable, though, is that the 10 hockey sticks presented in MM05 (one of the papers, you probably know which one) were not selected randomly from their sample of 10000. They were chosen to be most hockey-stick like. People, however, clearly interpret the results of MM05 as implying that random red noise typically produces hockey sticks, rather than random red noise sometimes (probably quite rarely) produces hockey sticks.

I promptly pointed out Anders was wrong, and what he claimed had been done hadn’t been done. He was attributing something which had been done in the Wegman Report, written by entirely different people, to Steve McIntyre and Ross McKitrick. This fact is immediately obvious to anyone who even looked at the paper in question (though this was complicated by virtue of Anders refusing to provide a specific citation).

Instead of citing the part of the paper where “the 10 hockey sticks [were] presented,” Anders cited some code for the paper, saying:

I actually have to go to a meeting so I’m going to not be responding anyway for a while. I really think you should read Nick Stoke’s post more carefully. I also, may have been wrong about it being 10, but here – I believe – is the bit of M&M05′s code in which the select the 100 most hockeystick like from their sample

This is where I made my mistake. I told him:

The code you referenced was not used in any McIntyre and McKitrick paper. It was code they wrote but did not publish the results of.

The problem is I didn’t notice Anders’s trick. I assumed he provided code which supported his original claim. He didn’t. The code he cited had nothing to do with his original claim.

His original claim held McIntyre and McKitrick presented cherry-picked graphs to make their results seem more certain than they were. His new claim was merely that McIntyre and McKitrick selected “the 100 most hockeystick” grahhs. Nobody had discussed that claim, and more importantly, nobody had disputed it.

I missed the change in topic. I assumed Anders was providing code relevant to what was being discussed, and knowing what he claimed was published had not been published, I assumed the code must have been extraneous. That was my mistake. I should have, as Steve McIntyre puts it, kept my eye on the pea under the thimble.

So what did the code Anders provided actually do, you ask? It picked the strongest examples of the effect McIntyre and McKitrick highlighted. Aha, you might think. That proves they cherry-picked their evidence!

No. As I discussed in a comment Anders deleted, McIntyre and McKitrick clearly described their “cherry-picked” graph as:

The simulations nearly always yielded PC1s with a hockey stick shape, some of which bore a quite remarkable similarity to the actual MBH98 temperature reconstruction – as shown by the example in Figure 1.A

The paper clearly stated the example was an example representative of only some of their simulations. The fact McIntyre and McKitrick “cherry-picked” the strongest examples to show what their strongest examples looked like is completely unremarkable. That’s exactly what they said they did.

In other words, Anders claimed McIntyre and McKitrick presented 10 deceptively chosen hockey stick graphs. When challenged on this, he presented code which didn’t select 10 hockey sticks and didn’t do anything deceptive. Instead, he presented code which did exactly what the paper said was done. It was a complete and total non-sequitur.

And I fell for it. I assumed what he provided must have had some relevance to the discussion, and as such, I responded incorrectly. That was a mistake, and I apologize for it.

Next time, I’ll try to remember not to assume what Anders provides has any bearing on what is actually being discussed.


One comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s