WHITEPAPER · SIMULATION STUDY

The SE You Buy Is the SE You Get: What 15,000 Simulated Certification Decisions Show About Severity Drift

Judge severity varies, and the field-standard fix, one average omnibus correction, cannot see where it lands. In a 15,000-draw Monte Carlo simulation over our own judge telemetry, measurement sold at a standard error of 0.20 was delivered at 1.16 with no correction, at 0.79 under the average correction, and at 0.21 under capability-level correction. MODEL OUTPUT PILOT ANCHORS

What is inside

The problem. Judge severity varies across capability levels, and one average correction cannot represent a bias that changes sign. The parameterization is wrong, so no amount of data fixes it.
This study. 15,000 simulated certification decisions over the measured severity curve from our own Nano judge probe, with the method appendix, every parameter published, and an honest claim tier: model output at proposed prices, never measured field fact.
The three-way contrast. No correction delivers a standard error of 1.16 against the 0.20 sold. The omnibus average leaves +1.34 of leniency on weak low-band work and invents -0.73 to -1.44 of severity on strong work, the sign-flip an average cannot see. Capability-level correction delivers 0.21 against the 0.20 spec at zero marginal cost, with zero mean residual in every capability cell.
The exposure math. At ten certified decisions per year at proposed flagship pricing, low-band mis-banding exposure runs $2.4M to $7.2M per year uncorrected, $0.85M to $2.6M under the average, and $0.34M to $1.0M under capability-level correction. Decision-theoretic exposure, not revenue.
The asymmetry a buyer should care about. A false pass admits a biased judge whose every later measure inherits the bias, a wave-scale contamination event. A false fail costs only the foregone judge-line saving. The risk sits orders of magnitude on the false-pass side, exactly where the average correction leaves +1.34 of leniency.
Traceability as the mechanism. How per-capability-level correction works, and why it makes every certification decision auditable after the fact, line by line.

The headline contrast lives on the differentiators page; this whitepaper is the depth document: the full simulation, the parameters, and the audit trail.

Get the full whitepaper

Get the full simulation: the method, the parameters, the three-way contrast the field-standard correction misses, and the audit trail that lets you check any certification decision line by line. Model output at proposed prices, parameters published, nothing withheld.

The PDF opens immediately. Six fields, one download, no drip sequence.

What you can do with it

Audit any certification decision we have ever made, line by line. The evidence trail is the product, not a promise about the product.
When a result is challenged, you do not defend a score. You show the record: what the judge saw, how severity was corrected at that performer's level, and why the decision stands.
If a performer was mis-banded, you see exactly where, at which capability level, by how much, and what the corrected decision would have been.

Every number on this page is model output from a simulation on our own judge telemetry at proposed prices, with pilot-scale empirical anchors. The simulation script, seed, and machine-readable results are published in the whitepaper. Nothing withheld.