Property scores: what a rating out of 100 really measures (and what to ask before trusting it)

“This property scores 61 out of 100.” Few things are as persuasive as a large number displayed in green or red: it condenses weeks of uncertainty into a single glance. That is precisely why it deserves more scrutiny than anything else in the report. A score may be a rigorous summary of verified data… or a decorative number that gives an opinion a veneer of science. From the outside, the two look identical. The difference lies in something hardly anyone shows you: the methodology.

What a score needs in order to mean anything

Any credible index — from an energy performance rating to a credit rating — publicly answers four questions. First: which variables are included in the calculation (risks? price? age? surroundings?). Second: how are they weighted — does being in a flood zone carry the same weight as having an F energy rating? Who chose those weights, and on what basis? Third: what is the benchmark — is 61 relative to every home in Spain, homes in that city, or homes at that price point? Without a reference sample, the number exists in a vacuum. And fourth: how has it been validated — does the score predict anything measurable (additional costs, depreciation, claims), or does it merely rank opinions?

If a service answers all four questions in publicly available documentation, its score is a legitimate tool whose limitations you can understand. If it does not, the number is not a measurement: it is an opinion dressed up with decimal places.

Three common tricks behind decorative scores

• False precision. A score of 61 (rather than 60 or “medium-low”) suggests laboratory-grade accuracy. But if the input variables are estimates — a property's condition inferred from a photo, or a special assessment that nobody has verified — the apparent precision is illusory: no calculation is more reliable than its weakest input.
• An average that hides the extreme. Overall scores average things out, and averages conceal serious issues. A property in a high-risk flood flow zone but with excellent surroundings might receive a passing score, even though that single factor could be reason enough not to buy. Deal-breakers should not be averaged away; they should be flagged.
• A negotiation anchor. An “objective” number shapes your judgement: if the report says 72, the property is “fine” and there is less incentive to negotiate. Always ask who benefits from that anchor, and remember that a score nobody can audit is an anchor nobody can challenge.

Why Hogarfax does not (yet) give an overall score

People ask us for one, and we understand why: a single number is simple. But an honest score requires exactly what it requires — verified variables, justified weightings, a meaningful comparison sample and validation against real outcomes — and we would rather publish no score than a decorative one. When our accumulated data supports a defensible index, we will launch it with the full methodology published alongside it, so that anyone can audit and challenge it. Until then, we show you the individual components: each check with its result, source and date. It is less eye-catching than a green 61. It is far more useful when making a decision.

The questions that expose any score

• Is the full methodology published, including variables, weightings and the comparison sample?
• Which input data has been verified and which is estimated? Does the score distinguish between them?
• Can a serious issue (high flood risk or an unfavourable ITE building inspection) be concealed by the average?
• Does checking the same property twice produce the same score? What happens if the source of one data point changes?
• What has the index been validated against: does it predict something measurable, or does it merely rank properties?

You do not need to be a statistician to use this list: simply send it to the service concerned and see how it responds. A sound methodology is something to show with pride. Silence, vague answers or a bare reference to a “proprietary algorithm” are also answers — and very revealing ones.