Safe Speed Forums

For some time Safespeed has claimed that the introduction of speed cameras has coincided with the loss of the previous trend in the reduction of road deaths.
This connection, AFAICT is unproven and based on the premise that if A and B are statistically related, A must be the cause of B (or B the cause of A). (This is a very dodgy assumption, but we will let that pass for now).
I would love to be proven wrong in this, but it occurs to me that there was another important change in road policy at more or less the time that speed cameras were introduced.
This was the wholesale cancellation of road building projects. It is well known that road safety is degraded when congestion increases - for example the M6 toll is much safer than the more congested main M6.
So, how do we know that the loss of trend is the result of speed cameras rather than the cutback in road building, and the failure of road capacity to keep pace with the increase in traffic?

I don't think the "loss of trend" claim is that the cameras have directly stopped the reduction, but that all the "other things" that used to be done before the reliance on cameras stopped.

The claim is that the obsession with the "speed kills" policy, has led to the focus on many other issues being lost & thus the loss of trend.

Personally, I believe that cameras can be dangerous and have no doubt in my own mind that some accidents have been caused directly by them, but as you say, there's no evidence of this.

It may well be that you're right, with the lack of road building & congestion. It's an interesting thought - slower roads more dangerous that faster ones!

_________________
Wake me up when the revolution starts
STOP the Toll Tax http://www.traveltax.org.uk

I would be very vary of comparing anything to the M6 Toll, since the high price and availability of alternative routes means that this road is very selective in terms of the kinds of drivers who use it. Ie. it will be limited to the rich, those in a serious hurry and anyone who can claim it back on expenses.

That said, the M6 Toll is a shining example of how speed doesn't kill! No wonder they started cracking down on speeding on it so quickly. Couldn't have a few years worth of meaningful statistics piling up now could they?

There's no fallacy of assumed causation, thankfully. The correlation came long after the effect was expected and functioned as confirmation, rather than original evidence. It fitted into a pattern.

I looked at road building budgets, and although I don't have figures to hand presently I recall that the spend was 'all over the place' and hard to match to any changes on the road.

TRL have investigated the effect of congestion on crashes, and in most circumstances found that crash risk fell with increasing congestion. This may be due to congestion naturally causing lower traffic speeds, or may be to do with increased 'traffic habituation' (e.g.. the greatest dangers come from the unexpected and the more constant the traffic the lower the risk that traffic will be unexpected.)

I estimated that in the past the total investment in roads was worth under 2% pa reducting in fatal crashes. There are a few papers (at least) that make the benefit 0%. I think black spot treatments and bypasses (especially) DO reduce crash risks overall, but the contribution is unlikely to be enough to account for a loss of trend of 5% pa or more.

TRL investigated the 'failure of fatalities to reduce' and concluded that 'drivers must be getting worse'.

So I actually rate it as highly unlikely that cancelled road building has caused the loss of trend.

If some small part of the loss of trend is due to underinvestment in roads, we should still properly blame the government for 'bad policy' that has failed to save lives.

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

This seems a bit suprising to me. One could argue that the "average" driver has a given rate of error over a given distance, or more likely time. So, the frequency of driver errors on a given stretch of road should be related to the number of drivers inhabiting it, everthing else being equal. The more drivers on the road, the more likely it is that one of them will make a mistake. One might expect that the accident rate would be related to the driver error rate. So more drivers on a given stretch of road should mean more accidents.

Anyway, I don't believe that you think that lower traffic speeds necessarily reduce accident rates enough to compensate for the effect I mention above (or at all - whatever TRL may have concluded ). But I am prepared to accept that a reduction in the RANGE of speeds in use could reduce both the driver error rate and he likelihood that an error will result in a crash. After all if all the vehicles on the road are travelling at exactly the same speed and in the same direction, they cannot collide. In congested conditions, speeds are more uniform as anyone can observe.

There's the whole "arousal" thing as Paul calls it. Basically with more traffic about, there are more things going on to keep the driver alert. Ploding down an empty dual carriageway is boring and the mind tends to wander, but when it's full of cars and people are doing random stupid lane changes and so on, suddenly negotiating it at 30 becomes a lot more interesting so you stay focussed.

In really high congestion, then everybody is moving at very slow speeds and is probably stationary for much of the time - it is no real surprise therefore that the accident rates drop. In total gridlock we would have zero speed and would expect zero accidents.