Failure run wild. …
I don't know any more about climatology than you do. And I'm guessing that you do not know anything about the CRU. Because apart from the name of the unit, it's address and that it has been in the bad news recently, all I know about it is that it hasn't accomplished anything useful.
I may strike you as fairly knowledgeable in some things about the way the planet works. But I am a semi literate amateur in geo physic circles. And some of the stuff I have discovered, although it works, is ignored by academics in office all over the material universe.
I am treated as a rank outsider in the premier newsgroup for British meteorology: uk.sci.weather.
Most of this is because of my ill-nature but a lot of it is that I keep on posting stuff that strikes those who have been to college and studied the art of weather forecasting (and made successful careers out of it too) as absolutely ridiculous.
Which is fair enough, my attitude has been a mirror of the behaviour of the CRU.
And the boulder I have been pushing uphill is about as acceptable an activity as looking at statistics.
But anyone who is looking at statistics to analyse how the weather works is a complete idiot.
What sort of analogy can I give, to help you to understand the idiocy?
Pick a card.
A deck of cards is designed with one face to be completely alike in every respect. But the other side is designed to be as different as possible.
Without a proven means of analysis, it is impossible to guess what suit, never mind what value the next card turned over will be.
But if you go through the deck once then repeat the process without shuffling the cards, anyone with a good memory will be able to guess what comes next. If they have a photographic memory or they took notes, they will know exactly which card will come next.
Keeping notes is another way of saying creating and using data.
All meteorologists keep data -from their own weather stations or from government funded ones. They love the stuff. When I went to school the geography teacher kept a weather station for the school and used it to teach final year students the art.
The magic has died with modern chilishness leaning towards vandalism. (Something that is copied in national politics here.) But with the internet, officially collected data has become widely available.
Oodles of it. And the CRU must have it coming out their ears.
But climate research still insists on looking at the wrong face of the cards.
Without understanding the root causes of weather, the systems produced by the engine that fires both the skies and the asthenosphere are blank faces in a shuffled deck of cards.
If a practiced poker hand is watching the way the cards fall in a game that has been going on all night, he will notice small signals in the way that a dealer shuffles the deck. And for short runs, may have some success in guessing what cards are going to follow certain other cards.
When you shuffle a deck of cards, the chances 50:50 that two adjacent cards will remain together or be separated by an unknown number of unknown cards. Sometimes an ace will remain adjacent to the card it lay next to during the previous hand.
That is the value of a climatologist's statistics.
An accomplished player will have a good idea about the odds of any of the 52 cards turning up in a way he can make money.
But he will never know exactly how the cards have been shuffled.
Climatologists hope to guess that the way the next hand is dealt is going to run like that. Seeing how one prvious game unfolded is no way to forecast how the next one will. But when they identify periods of similarities they might guess how the next few cards will tend to fall.
And that is about as far as old data can take you unless you spot a fault in the shuffle. Someone who can see those faults can "break the bank". Such a thing happened in Monte Carlo. Someone noticed that the roulette wheel on the table he was on had a bias.
Once he realised the way the numbers tended to come up he had a winning system. He was playing with someone else's "loaded dice".
With cards and with dice too -though less often, a visible bias can creep in as the cards get worn or the dice damaged. And so they are regularly replaced in casinos.
And these days the roulette table is checked daily.
With each phase of the moon a "new deck" is introduced.
Without taking that into account, no climatologist will ever get to know the run of the weather better than 6 or 7 days. And the chances of them getting good results past 3 to 5 days is always going to be fairly poor.
Better than 50:50 maybe -but not startlingly so.