The Survivorship Problem

Almost all advice about success is assembled from people who succeeded, which means it describes what winners did and not what caused winning. For teams that want a concrete workplace example, additional context.

The distinction matters because the losers did many of the same things.

The problem, plainly

A study of ten winning entries finds common features.

What it cannot find is whether those features appeared equally in the hundred that lost, because nobody collected them. For another perspective on recognition, evidence, or professional standards, see Behance.

Which means a list of "what winning entries have in common" may be a list of what all entries have in common, and the reader cannot tell.

The same applies to every genre of success writing — including the material this domain published before, which was survivorship at its most concentrated: accounts of people who arrived somewhere, with the arrival treated as evidence about the route.

Where it bites in this subject

"Winning entries tell a compelling story." So do losing ones. The observation is untestable from the winners alone.

"Successful applicants build a relationship with the funder." Unsuccessful ones frequently do too, and the ones who did not may have been in a different category.

And case studies of winners, which are the commonest form of awards advice and are almost entirely survivorship — useful as illustrations and not as evidence.

What escapes the problem

Structural facts about the process.

That an ineligible entry is rejected before scoring is not an inference from winners; it is how the process works.

That an unanswered criterion scores zero follows from the arithmetic.

That a criterion weighted at forty per cent is worth four times one at ten is definitional.

This site rests on that kind of claim wherever it can, precisely because survivorship does not touch it.

Where this site is exposed

Wherever it says what strong entries do.

"Evidence rather than adjectives" is the central practical claim here, and the honest status of it is: organisers and assessors consistently report it, it follows from how scoring works, and it has not been tested against a controlled sample of losing entries.

Which is stated on the limits page rather than buried.

The claim is not weak. It has two independent supports — mechanism and practitioner report — and neither is a controlled comparison, which is what a reader should know.

What a reader should do about it

Weight structural claims above observational ones.

And treat any "what winners do" list, including any on this site, as a hypothesis rather than a finding.

The useful test is whether the claim explains a mechanism. "Judges score criteria, so answer them directly" explains one. "Winners are passionate about their work" does not, and it is the shape of advice that survivorship produces most reliably.

The version that appears in awards guidance

Frequently and in a recognisable form.

"We analysed a hundred winning entries and found..." — which is the problem stated as a methodology.

Where such an analysis also examined losing entries, it is genuinely informative and says so, because the comparison is the whole value.

Most do not, because organisers hold the losing entries and rarely release them, and consultancies see only their own clients' work.

Read for the comparison group. Its absence is not dishonesty and it is the thing that determines what the finding means.

Why the problem is hard to fix here

The data exists and nobody can publish it.

Organisers hold every entry, winning and losing, scored against the same criteria — which is exactly the dataset that would settle most questions on this site.

Confidentiality prevents it. Entrants supply commercially sensitive material on the understanding that it stays with the panel, and no organiser is going to publish a comparison of strong and weak submissions.

Which means the honest position is permanent rather than awaiting better evidence: this subject will continue to rest on mechanism and practitioner report, and anybody claiming otherwise is claiming access nobody has.

The mirror problem

Failure narratives have it too, in reverse.

"Why entries fail" assembled from rejected entries describes what rejected entries did — and the winners may have done much of it as well.

The exception is administrative rejection, which is recorded and countable: an entry ruled ineligible was ruled ineligible, and no inference is required.

Which is why this site leans hardest on that section and states the rest more carefully.

What would settle it

A programme publishing anonymised scores across all entries in a category, with the criteria attached.

One category, one year, would answer more than every awards guide in existence — whether evidenced entries score higher, whether length correlates with score, whether administrative failures cluster.

A few research bodies have done comparable work in grant assessment, where funders occasionally release anonymised data.

Nothing equivalent exists for awards, and that absence is worth stating rather than working around.

The practical consequence for a reader

Not paralysis.

Advice resting on mechanism is reliable enough to act on, and it happens to be most of the actionable material here: answer the criteria, supply evidence, submit early.

Advice resting on observation is worth trying and not worth agonising over — and where it conflicts with your own experience, your experience is the more specific evidence.

The distinction takes a moment to apply and it prevents the two failure modes: dismissing everything because none of it is proven, and following a case study because somebody won.

The short version

• Advice assembled from winners describes what winners did, not what caused winning, because the losers were never collected
• A list of what winning entries have in common may be a list of what all entries have in common
• Case studies of winners are the commonest form of awards advice and are almost entirely survivorship
• Structural facts escape the problem: eligibility, scoring arithmetic, and weighting are definitional rather than inferred
• This site is exposed wherever it says what strong entries do, including its central claim about evidence over adjectives
• Weight structural claims above observational ones, and test any claim by whether it explains a mechanism