Past performance, honestly: what calibration adds to a track record
Every track record is a story the manager tells. Diligence is deciding which stories to believe. Calibration is a second axis the returns alone can’t give you — was the stated conviction honest?
“Past performance is no guarantee of future results” is on every document precisely because it’s true: a track record is a story, and a good story can be told by skill or by luck. Allocators spend their lives trying to tell the two apart. Calibration is a tool for that — a second axis the return series alone can’t give you.
Returns underdetermine skill
A manager who was right four times running might be brilliant or might be a coin that landed heads four times. Returns alone can’t distinguish them, especially over the short, noisy windows most allocations actually have. You need something that probes the process, not just the outcome.
What calibration adds
Calibration scores the conviction, not just the result. If a manager records their forecasts with probabilities — “70% this holds,” “60% this re-rates” — those probabilities can be scored against what actually happened, with a proper scoring rule. A well-calibrated manager’s 70%s come in around 70% of the time. That’s a fingerprint of genuine edge that’s much harder to fake than a return number, because it’s evaluated claim by claim over many calls.
A credential you don’t take on faith
The point of turning that into a Calibration Attestation is that your diligence team doesn’t have to believe it — they re-derive it. The forecast history is scored, hash-locked, and recomputable in your own environment: a track record expressed as something checkable rather than a backtest you’re asked to trust. (Honest caveat: the signature today is ours, recomputable by you; an independent anchor is in progress.)
The asymmetry
For a manager, a calibration credential is a way to stand out from everyone telling the same confident story. For an allocator, it’s a cheaper, faster filter than another month of reference calls. Both sides win for the same reason the rest of AEQUARA works: a claim you can recompute beats a claim you have to believe.