Dear friend,

I remember at last year’s Monte Carlo one of the keenest observations (I think it came from an Aon exec) was that we had been really lucky that “a hard market for reinsurance had coincided with a soft market for hedge funds”.

Well, the soft market for hedge funds is well and truly over.

Fools bought collateralised debt obligations based on loans made to people who couldn’t pay them back collateralised against hopelessly overvalued housing stock .

And since these things were so complicated, and they were illiquid, they marked them ‘to model’ and not to market. Some of these models thought that the events of the last week were ‘25-sigma’ ones – meaning that they are supposed to happen once every 100,000 years.

How funny then that many separate one-in-hundred thousand year events happened day after day over the last couple of weeks!

Any of this stuff sound familiar?

Are we ever going to learn that you can't let a mathematician run your business for you? How many Nobel prizes did they have at Long term Capital Management - and what the hell good did it do them when they were stuck with a load of junk that non-one wanted to buy (but which the model said was theoretically worth something?

Markets are markets - and maths is only a helpful tool when things are normal. When irrational fear or exuberence stalk the streets, mathematical theory is about as useful as an umbrella made of papier maché.

Back in 2005 and 2006 we were lucky it was so easy to reload the capital barrel and start cranking the handle again.

But now it might not be so easy.

Hang on to that pile of cash marked 'jumbo 2006 profits' for now, it might come in handy.