Safe Speed Forums

It'd be good to know more about the distribution of repeat speeding offences. I expect them to be essentially random - i.e. Anyone might get a speeding conviction and anyone might get two or three.

The Direct Line survey published today and widely reported allowed me to calculate the following figures:

The poll found that 16% of drivers (5.4 million drivers) have points on their
driving licences, with 2.2% of drivers on 6 points (750,000 drivers) and 0.48%
of drivers (160,000) on 9 licence points.

There are a number of reasons that the figures are not precise, but the pattern closely matches the random distribution theory.

Suppose we throw balls at random at 1,000 buckets until 20% of the buckets have at least one ball. How many of the buckets will have 2 balls, 3 balls, 4 balls and so on?

I reckon it's:

1 or more: 20%
2 or more: 20% of 20% = 4%
3 or more: 20% of 20% of 20% = 0.8%

This is very much the pattern that the Direct Line figures suggest. i.e. the Direct Line figures are consistent with random distribution of speeding convictions.

There are 'confounders'. Firstly behaviour could change to make further points less likely, for example:

- someone who has 6 points may avoid speeding and might therefore expect to receive fewer points in the future.
- someone who has 6 or more points may palm further points off to someone else to avoid the risk of being banned.

Secondly points might be given to a smaller groups of 'serious speeders' within the population.

I'm quite sure that all these effects are present within the sample - but isn't it interesting that the results are pretty consistent with random distribution.

"I'm sorry pal, your number came up. Please pay £60."

_________________
Paul Smith
Our scrap speed cameras petition got over 28,000 sigs
The Safe Speed campaign demands a return to intelligent road safety

The distribution follows the kind of pattern you would expect, but you can't really extrapolate that the chances of getting a speeding conviction are essentially random. There are various factors that influence the chances, such as where you live, the total mileage you do, and the proportion of your driving done on unfamiliar roads.

Also being aware of what the speed limit actually is, and having some knowledge of enforcement techniques, helps avoid convictions. I still firmly believe 80% of the motoring population haven't got a clue what a Talivan looks like.

Obviously saying "if you don't speed, you won't get caught" is far too simplistic, but neither can you say that each motorist stands anything like an equal chance of being caught in each 100 miles driven.

_________________
"Show me someone who says that they have never exceeded a speed limit, and I'll show you a liar, or a menace." (Austin Williams - Director, Transport Research Group)

Any views expressed in this post are personal opinions and may not represent the views of Safe Speed

Those factors don't alter the 'essential randomness' that I was trying to reach down towards - because at the system level they are automatically equalised.

We can never assess the risk to a particular individual - but we can assess the risk experienced by an average individual. We may then be able to further sub-classify individuals by group or type and then assess the risk within the average group or type member.

In this case the ultimate question is 'can speed cameras identify risky drivers?' and the answer is (generally) of course not.

_________________
Paul Smith
Our scrap speed cameras petition got over 28,000 sigs
The Safe Speed campaign demands a return to intelligent road safety

I think we're basically in agreement, but you can probably assign to each driver an individual "speed camera risk profile" based on various factors, many of which bear no relation to actual risk on the roads.

It may well be the case that avoidance of prosecution is increasingly dependent on "speed enforcement awareness" which doesn't necessarily correlate with safe driving, or indeed speed limit compliance.

_________________
"Show me someone who says that they have never exceeded a speed limit, and I'll show you a liar, or a menace." (Austin Williams - Director, Transport Research Group)

Any views expressed in this post are personal opinions and may not represent the views of Safe Speed

Agreed - although you have moved well (and interestingly) away from my original point.

You have also helped to to reframe the original idea so that I'm able to move towards saying that:

The distribution of repeat speeding offences does not contain information to support either of the following propositions:

a) There is a significant group of risky repeat offenders
b) Speeding convictions alter speeding behaviour

(just to making it clear - I'm NOT yet ready or able to make that statement - but I'm getting closer)

_________________
Paul Smith
Our scrap speed cameras petition got over 28,000 sigs
The Safe Speed campaign demands a return to intelligent road safety