How Stressline scoring works
Stressline scores measure structural overlap between your thesis and 241 documented healthcare startup failures across 15 sub-verticals. Higher scores mean your structural profile more closely resembles companies that failed in the same failure modes. The score is not a prediction — it measures how much structural evidence exists against your thesis.
What the score measures
The score aggregates three structural signals:
Corpus match strength
How many of the 241 coded failures share your thesis's structural dimensions — buyer type, mechanism, and status quo disruption. Computed via embedding proximity across the full corpus, not keyword matching.
Sub-vertical density
How concentrated the failure history is in your sub-vertical. A sub-vertical with 33 coded failures gives sharper signal than one with a handful.
Failure node concentration
Which structural failure patterns dominate your sub-vertical. When Outspent by bigger competitors accounts for a large share of failures in your space, it raises the structural risk score.
Evidence depth tiers
The score is accompanied by an evidence depth classification that tells you how much structural precedent exists for your particular thesis.
The corpus has substantial precedent for your thesis dimensions. The prosecution case is built on solid structural evidence. Take the findings seriously.
Enough precedent to identify structural risks, but the pattern is not statistically saturated. Findings are directional, not definitive.
Limited precedent in this sub-vertical or structural profile. The corpus lacks sufficient density to make strong structural claims. Treat findings as early signal only.
How scores map to verdicts
The structural risk score feeds a Bayesian posterior that determines the final verdict classification. Three thresholds partition the score range:
KILLED — the structural evidence against your thesis is strong enough that the primary failure mode has dominated every structurally similar company in the corpus. This verdict requires both a high posterior probability and a minimum evidence depth of MODERATE or above. PASSED WITH CONDITIONS — the thesis contains identifiable structural risks but structural survivors exist with similar profiles. The conditions specify which dimensions require differentiation. PASSED — insufficient structural overlap with known failure modes. Not a guarantee — it means the corpus does not have strong precedent against this particular configuration. Thin sub-verticals can produce PASSED verdicts by default; the evidence depth tier will indicate this.
What the score does NOT measure
This is not a prediction. The Stressline score is deliberately not framed as a survival probability. Here is what it cannot tell you:
- —
Not a survival probability. A high score means strong structural pattern overlap — not a guaranteed failure.
- —
Not investment advice. The score is a diagnostic tool for founders. It does not account for team quality, market timing, or execution.
- —
Not a prediction of which failure mode will occur. The corpus identifies the most common structural risk. Other risks may be equally important.
- —
Not a statement about your specific market. It reflects the structural history of similar companies, not your specific competitive situation.
Corpus basis: Scored against 241 Gioia-coded failures across 15 sub-verticals. Each entry independently verified with a 5-signal death composite (domain, LinkedIn dispersal, Crunchbase, press coverage, funding cessation). Minimum 3 signals required for inclusion. See corpus methodology for full verification protocol.
Related methodology