In a remote house in the middle of Dartmoor six shadowy figures huddle around a small round table for a séance. Tension rises as the spirits spell out a chilling message: 'Captain Trevelyan...dead...murder'. Is this black magic? The only way to be certain is to locate Captain Trevelyan. His home is six miles away and with snow drifts blocking the roads someone will have to make the journey on foot.

The year is 1931 and the author of The Sittaford Mystery is a young writer of detective stories. American aficionados may know the book by the title from its New York publisher Dodd & Mead…Murder at Hazelmoor’s. It was the first novel by Agatha Christie to incorporate the supernatural.

Agatha Christie describes Sittaford as ‘a small remote village on Dartmoor under the shadow of Sittaford Beacon.’ ‘…they went upwards over a rough moorland road until they reached a village that was situated right on the edge of the moor. It consisted of a smithy and a combined post-office and sweet shop. From here they followed a lane and came to a row of newly-built granite bungalows…’

In the book Sittaford is six miles from Exhampton whose features match Okehampton pretty well. For Sittaford read Throwleigh…with allowance for the odd piece of artistic licence.

The trouble with nice places is hills. They look pretty but it rains a lot. But Science has the answer. It is not the rain per se that causes the trouble but being underneath it as it accelerates earthwards. With the same rain some people get wetter than others. Why?

If the children are driving you mad over Christmas then why not turn them into Gentlemen Scientists? Here is how. Mark out a 500-yard circuit around the garden and while the children are asleep weigh their clothes. In the morning when it is pouring with rain and the little brats are yelling the house down collar the one who usually gets the blame and make the little dear run the 500-yard circuit in the rain. Tell the rest to time it…or else.

By this stage the real culprit will be gloating. This is next one to send out into the rain…with the instructions to walk and not run the same circuit. Finally get them all involved in weighing the wet clothes. If more punishments are needed have at the ready a supply of toppers, bowlers…and the hat your mother wore for her wedding.

By now you will have discovered that the Walker’s clothes were 40% wetter than the Runner’s…plus or minus for the direction of the rain and the shape of the experimental object. It is a long holiday. Repeat the experiment as often as necessary to establish the relationship between wetness and (a) the shape ‘n pace of your test object and (b) the slant ‘n speed of your rain.

Your working hypothesis is that a fat person gets wetter than a skinny one…and the head and shoulders get wetter with rain that comes straight down. For a Doctorate in Displacement Activity (PhDA) find out the parts of the body that get wettest. To check your results click here for the frontiers of Experimental Science.

Should you wish to plan your Holiday Displacement Activities you can always phone the Met Office. But I should perhaps dampen your expectations. What you want to know is whether it will rain tomorrow. What you will be told is something more along these lines. ‘Over much of the European Region the situation is now finely balanced with approximately even chance that the winter will be colder or warmer than average.’ Douglas Adams almost got there before me…but I will claim it as Shepherd’s Law. ‘The more expensive the model the less useful its output’.

Forecasting Future Economic Behaviour is often taken as the proper test of an Economic Model but forecasting future conditions is a losing game…and similar considerations apply to past data. The near future is an extrapolation of the past because nothing can quickly divert the path. But the likely future range of the path diverges with time due to randomness and the dynamics inherent in the structure of the system.

The time horizon for an effective forecast is that interval during which little deviation is possible from future disturbances. Action taken at the decision time will have little effect in the forecast time horizon because it takes time for the effect of a policy change to affect the system. This means that any action at the decision time will be effective mostly in the action region…which lies beyond the forecast region.

So the usefulness of forecasting is limited by the relationship between the short time within which to make a reliable forecast and the long time required to affect the system. We can call this Forrester’s Law after the MIT Professor who discovered it.

‘One can forecast in the time zone in which one cannot act…and can act in the time zone in which one cannot forecast’. This does not mean models are useless. But effective models are different. They show how the nature of the behaviour of a system would be altered by consistently following an alternative policy.