UKC

Car insurance algorithms

New Topic
This topic has been archived, and won't accept reply postings.
 sheelba 26 Oct 2016
Hoping that someone on the forum might be able to shed some light on the murky world of car insurance algorithms. I recently had to change the address of where my car was kept to Settle in North Yorkshire. I've heard that traffic density is one of the main determiners of price so hoped that the premium would go down, instead it went up. This has been rectified now with the company however using another companies website it was a whole £260 more expensive to insure the car in Settle than it was at my parents home in Lymm near Warrington (choosing houses of a similar value and keeping all other variable constant). I'm struggling to understand why this is. Looking at vehicle crime in settle it does not seem to be particularly high. It is in the Bradford post code which I know has high prices due to a lot of vehicle crime but risk checkers seem to suggest that Settle (BD24) is low risk unlike the centre of Bradford. I know the A65 is an accident blackspot is this driving it up?
 balmybaldwin 26 Oct 2016
In reply to sheelba:
I can't comment directly without knowledge of the systems used by the car insurance company involved, but the following may shed light on it. (For information I work in data analytics in the insurance industry)

Firstly Bradford is in the top 5 postal areas in the country for motor insurance fraud (at least it is in my data). It is also among the highest areas for "propensity to claim" (either that or Bradfordians develop weak necks - maybe there's something in the water). Whilst postcodes are quite a coarse way of determining this, many insurance companies work at whole postal area level as their data isn't good enough to be more specific. Both these factors have a very severe urban bias and therefore a company with more advanced data gathering may be able to offer you a better rate for a more rural area within a high rate overall postal area.

Another factor is uninsured and unlicensed/provisional license driving this seems to be a particular problem in Bradford, Blackpool and across the rural north east, as well as London and Birmingham.

An insurer's risk appetite can often make big differences in the quotes that are offered for the same risk location regardless of other factors (and can often mean changing location mid-term can lead to very big rises). Some insurers (like the one I work for) choose to quote less for a rural postal area because they are after rural customers as in general they have a better accident record (although they do tend to have bigger accidents as rural roads are faster than urban ones). Others prefer the more predictable urban motorist as whilst the claims are more common they tend to be cheaper (although this is changing with increased levels of Low Velocity Bodily Injury claims being made without the need to prove injury, increased propensity to claim and higher levels of fraud.

Accident black spots will affect their general area, but I'd be surprised if the A65 has a particular problem - it's certainly not one notorious enough to have been discussed (unlike the A34, or the Cat & Fiddle used to be). The ones that spring to mind are the high passes (due to ice risk in the winter) and roads with specifically high accident:traffic flow ratio (just because a busy road has a lot of accidents doesn't make it high risk if the flow rate is high enough).
Post edited at 19:30
OP sheelba 26 Oct 2016
In reply to balmybaldwin:

Thanks, that's interesting stuff

New Topic
This topic has been archived, and won't accept reply postings.
Loading Notifications...