House Price Statistics – One Million Anecdotes

House Price Statistics – One Million Anecdotes

By Charlie Lamdin – Founder, BestAgent

This morning, sitting down at my desk, I noticed something new, and really weird.

While news headlines report house price falls, buyers withdrawing due to spiking mortgage rates and a general armageddon-style outlook for home movers, I spotted something I hadn’t seen before, in all my 24 years in property.

Looking at the BestAgent data, I could see properties coming back on to the market in unusually large numbers (no surprise given the news), having had a deal fall through, for whatever reason, but something was different. Usually, when a property is re-listed, it’s either at the same price as before, or, for obvious reasons, a reduced price.

Today, a noticeable number of these fall-throughs were being re-listed with their asking prices now 5-10% higher than before!

What? Sorry, what?

“What’s going on?” I consulted a few of my agent contacts.

“It’s so that they can offset the effects of all the down valuations and below-asking price offers they’re expecting now.”

In 24 years working in this industry, this is a new one to me.

It’s not because they don’t think prices are falling. They know prices are falling. They think that just by raising the asking price, they’re still going to get the price they actually want.

The madness of movers.

This got me thinking about how often I hear extraordinary stories surrounding individual properties, and how they’re always, invariably, unique.

The BestAgent algorithm runs 24/7/365, it never sleeps, always open for business. It constantly scans for all available properties For Sale and To Let, currently across England and Wales, (eventually it’ll do the whole of the UK).

It’s been running for a few years in preparation for coming to market, and we’ve been trying to make sense of the monolithic mess of online property data.

Early on in the build process, I asked the development team (crazy, brilliant geniuses), to create some kind of dashboard where I could see it working. (Although I’m the software architect of the entire BestAgent “Universe” as we call it, I couldn’t write a line of code if my life depended on it.)

The next day I had a screen that looked like something out of The Matrix, or Minority Report, with a never ending scroll of facts, figures, data and statuses that move too fast to read, unsurprising given it runs checks on about one million properties per day.

The property data set is almost useless. But after years of watching it, I’ve learned to understand it, and notice changes. It’s why I saw this house price fall coming back in February.

Fake listings.

It’s so full of fake, duplicated, out-of-date property listings, not to mention the listings that aren’t even homes (garages, land, shops, mobile holiday homes). We estimate that around one-third of all property being advertised has no right being there at all.

This is no surprise. Agents have been “gaming” the system to look bigger and busier than they are, and to attract mover leads on fake properties, to then try and show them their other properties.

But the problem is, it makes the property price data releases from the big well known property websites, almost useless.

One of our Supplier Partners from the house data sector once told me over a few beers that the bar of national property market data is “unbelievably low”, even from the respected indices. When he saw BestAgent and how we’d built it, he said “you’re going to be able to raise the bar easily with this.”

I wish!

We’re interested in homes, not investments, land, off plan etc, just homes. We try to strip out all the property listings that aren’t homes, available to view now, for qualified movers.

The data quality is so bad.

Then I had a thought: every single property listing, real or fake, is unique. Each one is some sort of exception. They are all an exception, in some way shape or form. And everyone is trying to get one over, be one step ahead, pull a fast one, to get the deal they want. Buyers, sellers, landlords, tenants and agents alike.

Homes aren’t in fact a commodity. They’re all unique, even in a street of identical terraced houses.

And this is one of the factors that makes interpreting property market price data so difficult, such an art form, so frustrating, so unreliable.

It’s one of the reasons why Automated Instant Valuation tools are completely useless. Totally, completely useless.

Even a seller’s reasons for moving are a factor in a property’s likely sale price – how can the algorithms price that in?

They can’t.

One million anecdotes.

And this is what makes the residential property market so completely, maddeningly, beautifully, different from any other market in the world. It’s why no one understands estate agents, and why estate agents behave so weirdly.

24 years in studying the whirlwind of emotional, irrational behaviour in this industry, and it still surprises me.

A million anecdotes. It’s the best summary of property market price data I’ve ever heard.