InstrumentsAvailableFor CIOs, quant PMs and allocators
How much of your Sharpe ratio is real?
Assay re-scores a backtest for the number of trials behind it, so you know how much of the result is skill and how much is selection, before an allocator asks.
“We tried 400 variations and promoted the best one. How much of that Sharpe is real?”
The question Assay answers.
02 · What you receive
A report you can forward to an allocator.
The headline result restated after the trials behind it, how much of it selection explains, whether it holds on data it never saw, and a verdict in one paragraph of plain English.
Momentum, top decile, monthly rebalance
Backtest supplied with 400 recorded variations
| Sharpe, as reported | 1.84 |
| Sharpe, after the trials behind it | 0.52 |
| Share of the result explained by selection | 71% |
| Holds out of sample | partly |
Most of the headline result comes from choosing the best of many tries. A smaller, real effect remains.
Layout specimen. Figures are illustrative, not a client result.
Four steps, in plain words.
Your backtest and the record of what was tried, including the versions you dropped.
We rebuild the result and confirm the record of trials is complete.
We score the result against everything that was tried to reach it.
You receive a short report written for an allocator, not for us.
It counts the tries you did not show, not only the one you did.
It is independent of the team that built the strategy, so the answer carries weight in diligence.
What we need from you
- The return series of the strategy you want assayed
- A record, or an honest count, of the variations tried
- A paragraph on what the strategy does
What stays with you
- Your code, signals and parameters
- Your raw data; it never leaves your environment
- Position-level detail, unless you choose to share it
Know the number before an allocator does.
Tell us what you are working on. A founder replies, and the first call settles scope. No prices, no packages; those come in the conversation.