What is the difference between a single-factor and a compound stress?

A single-factor stress moves one variable — the value, the works, the rate or the timeline — and leaves the rest at the central case. A compound stress moves several at once. Only single-factor losses cap the Deal Score; compound scenarios are shown but do not cap.

Why not just reject anything that loses money under stress?

Because almost every sound deal loses money under a severe enough compound scenario, so that rule would reject everything. The distinction that matters is whether the deal is fragile to a single realistic movement, and that is the one the engine uses to cap.

Which variable matters most?

Usually the works budget, because it has the widest error bar and feeds straight into true discount and the capital stack. A deal that only works if the works come in exactly on estimate is a fragile deal, and the Red Team says so.

Red Team stress testing: finding the fragile deal

2026-10-20·4 min read

The Red Team runs nine stress scenarios against every deal, moving the finished value, the works cost, the interest rate and the time to sell — first one at a time, then several together. Only a single-factor loss caps the Deal Score. A deal that only loses money when four things go wrong at once is a real risk and a bad reason to reject an otherwise sound purchase.

Every appraisal is a forecast, and every forecast is wrong. The useful question is not whether the central case holds but how much has to go against you before the deal stops working — and whether the things that would have to go wrong are one thing or four things at once. Treating those two cases identically is how a scoring system becomes useless.

What does the Red Team actually move?

The variables with the widest error bars and the largest effect: the gross development value, the works budget, the cost of finance, and the time the property sits before it sells. Each is moved against the deal, and the appraisal is recomputed in full — the same engine, the same true discount, the same after-tax margin, on worse inputs.

Single-factor stresses, and what each one tests

Finished value falls: Was the gross development value optimistic?
Works overrun: Is the contingency real, or a note?
Rate rises: Does the capital stack still close on dearer money?
Sale is slow: Does the deal survive a longer, costlier hold?

Why do only single-factor losses cap the score?

Because fragility and risk are different things. A deal that turns a loss when one variable moves a realistic amount is fragile: it is a bet that nothing goes wrong, and something always goes wrong. A deal that only turns a loss when the market falls and the works overrun and the rate rises and the sale drags, all together, is not fragile — it is a sound deal facing a bad year, and rejecting it on a compound scenario would reject almost everything worth doing.

So the tiering is deliberate and it is pinned by a test, because it is exactly the kind of nuance a later simplification would flatten. Single-factor losses cap the Deal Score. Compound scenarios are reported — a reader is entitled to see them — but they do not cap, because a score that treats a once-in-a-decade confluence like a base case is a score that never approves anything.

How is this different from a sensitivity table?

A sensitivity table shows you numbers moving and leaves the judgement to you. The Red Team makes the judgement structural: it decides which movements are allowed to cap a deal and which are not, and it applies that decision the same way every time. The output is not a grid to interpret but a verdict with its reasoning — this deal is fragile to the works budget, that one is robust across every single-factor stress and only fails on a compound worst case.

What it means in practice

A thin-margin deal that survives every single-factor stress can outscore a fat-margin deal that fails one.
The works budget is usually the variable that bites, which is why the contingency is charged into the deployed figure rather than noted beside it.
A deal resting on an optimistic finished value is caught here even when the headline true discount looks comfortable.
Resilience is a Deal Score component in its own right, so a robust deal is rewarded, not merely spared.

The point of stress testing is not to frighten the user out of every deal. It is to tell the fragile ones apart from the robust ones, so that the robust thin-margin purchase gets taken and the fragile fat-margin one gets left — which is the opposite of what a headline number does.

Common questions

What is the difference between a single-factor and a compound stress?: A single-factor stress moves one variable — the value, the works, the rate or the timeline — and leaves the rest at the central case. A compound stress moves several at once. Only single-factor losses cap the Deal Score; compound scenarios are shown but do not cap.
Why not just reject anything that loses money under stress?: Because almost every sound deal loses money under a severe enough compound scenario, so that rule would reject everything. The distinction that matters is whether the deal is fragile to a single realistic movement, and that is the one the engine uses to cap.
Which variable matters most?: Usually the works budget, because it has the widest error bar and feeds straight into true discount and the capital stack. A deal that only works if the works come in exactly on estimate is a fragile deal, and the Red Team says so.

Sources

Price Paid Data
Every registered residential sale in England and Wales since 1995. Published by HM Land Registry; the link goes to their site rather than to the document, because a deep link into guidance is a slug that changes without notice.
UK House Price Index
Official movement in values, by local authority and property type. Published by HM Land Registry; the link goes to their site rather than to the document, because a deep link into guidance is a slug that changes without notice.
Corporation Tax rates and reliefs
The main rate, the small profits rate and marginal relief between them. Published by HM Revenue & Customs; the link goes to their site rather than to the document, because a deep link into guidance is a slug that changes without notice.
Stamp Duty Land Tax rates
The bands and rates in force, and the reliefs that alter them. Published by HM Revenue & Customs; the link goes to their site rather than to the document, because a deep link into guidance is a slug that changes without notice.

Terms used on this page

true discount: Total money deployed measured against open market value, rather than the headline discount on the purchase price.
Deal Score: A 0-100 score computed from nine weighted components, always on profit after tax, never before it.
gross development value: What a property is expected to be worth once the planned works are finished.
capital stack: Every source of money in a purchase, ordered by who is repaid first and what each layer costs.
Red Team: Nine stress scenarios run against every deal, with single-factor losses treated more seriously than compound ones.