One book. Proven on paper, before a word of promise.
Canli Capital is a market-neutral (positioned to profit from relative moves between strategies, not market direction), multi-strategy book you reach through its live paper track record. Here is what it is, why we run it on paper first, and the transparency that governs every figure.
Decorrelated sleeves, one book.
ALPHAC is a market-neutral, multi-strategy book. Its core runs several deliberately different sleeves, each with its own economic premise and clock. They are combined to test breadth, not to manufacture a stronger headline.
The reason to hold them together is a falsifiable diversification thesis: losses in one stream should not automatically propagate through the rest. The research window is encouraging, but live-forward correlation is not established and the public evidence keeps that distinction visible. A disclosed strategic net-long overlay sits beside the neutral core as a separate, labelled line, never blended in, and it is stated plainly as commoditized beta rather than alpha.
The exact research allocation, the standalone Sharpe per sleeve, and the full equity curve live on the dashboard, read straight from the same source as this page.
Earned numbers, not flattering ones.
A backtest can be tuned until it shines. The discipline is to deflate it. The in-sample Sharpe is the headline figure; after correcting for multiple testing via deflation, the honest forward expectation is materially lower, with a real chance of roughly zero in year one. We present the deflated figure, never the flattering one.
The book has been put through stress tests and earned a grade of C plus: real but modest, and it fails its own multiple-testing deflation in-sample. The right response to a modest edge is to prove it live, not to dress it up.
So the access point is the paper track record itself. It begins at day zero with no fabricated history, and it accrues in front of you. When the live record speaks, it will have earned it.
What stays true.
- /01
The research curve is a simulation, not realised trading. No real capital has been deployed.
- /02
The live paper track record is shown exactly as it accrues. We publish no return until the record earns one.
- /03
Published states are linked and signed, selected chain heads receive OpenTimestamps checkpoints, and new entries disclose their canonical payload for independent rehashing. Earlier opaque commitments remain explicitly limited.
- /04
We always lead with the deflated forward figure, never the in-sample one, and we publish the dead ideas with their real net-of-cost numbers.
- /05
Historical simulations contain a modest diversified signal, but they do not clear our own deflated-Sharpe deploy gate and have not established an edge forward.
Step into the record.
Public access opens the dashboard: the metrics, the research simulation against the live day-zero marker, the sleeves, and the four transparency statements above, all read live from one source. There is no highlight reel, because there is nothing to curate yet.