Performance

Live track record

Data as of —, updated — UTC · published automatically, nightly

Validation (paper)

USD

Live

USD

Live trading with my own capital is planned for late 2026. This group activates on day one.

Total system

Indexed, start = 100 · USD, translated from the CHF account at the daily USD/CHF rate
Total system
Pre-registered expectation band, frozen before the validation phase started. How this band was derived →
No benchmark line: the account combines a trend programme with a passive ETF overlay and has no natural peer. The trend sleeve is compared with a managed-futures ETF proxy below.
Shaded region: paper validation.
scroll to zoom · drag to pan · double-click to reset
Drawdown · total system, USD

Monthly returns

Total system, USD, % per month
Shaded months: paper validation.

Trend sleeve vs. managed-futures ETF proxy

Both indexed, start = 100 · USD · sleeve on the full account value · proxy scaled to the sleeve's volatility target
Trend sleeve MF ETF proxy: DBMF + KMLM, equal weight, return above cash scaled ×1.65 (USD)

The proxy is an equal-weight basket of two listed managed-futures ETFs, DBMF (iMGP DBi Managed Futures Strategy) and KMLM (KFA Mount Lucas Managed Futures Index Strategy), rebalanced daily, total return with distributions reinvested (adjusted close, Yahoo Finance). Two products with fund expenses of roughly 0.9% p.a. and their own tracking error, not a peer-group index; managed-futures indices such as the SG Trend Index are licensed products and are not shown. The trend sleeve line is the result of the futures positions measured on the full account value, the capital the programme is sized on. The ETFs target a lower volatility than the sleeve. To put both lines on the same risk, the basket's daily return above the 3-month US Treasury bill rate is multiplied by a fixed factor of 1.65. The factor is the sleeve's volatility target divided by the basket's volatility over its whole joint history before the track began, December 2020 to 4 August 2026. It was set once and is not adjusted to the track. The bill rate is taken out first because the ETFs earn it on their collateral, and a multiplied line would multiply that income too; the sleeve line likewise contains position results only, no interest. Nobody can buy the scaled basket; it is a yardstick. Until 2 October 2026 this chart showed the unscaled basket and measured the sleeve on half the account, which doubled its percentage moves. Context, not a like-for-like race.

Methodology notes

The system trades a diversified universe of roughly one hundred futures markets spanning equity indices, government bonds, short rates, currencies, metals, energy and agriculture. Signals combine classic trend following across several speeds with carry and a small set of complementary systematic families, computed once per day on end-of-day settlement prices. Position sizing targets a constant portfolio volatility; a dynamic optimiser translates targets into whole contracts under explicit cost and margin penalties. The futures programme is sized on the full account value. On top of that, half of the account is invested in a passive overlay of world equity and gold ETFs at a fixed 70/30 split, unhedged; the margin efficiency of futures leaves enough cash to carry both, for roughly 150 percent total deployment. All figures on this page derive from the broker-reported account value in CHF, indexed to 100 at the close before the first trade. The USD track translates that value at the daily USD/CHF rate and therefore contains currency translation that is not part of the strategy. Nothing is converted in the browser. The configuration was frozen and the expectation band preregistered before validation began. During the validation phase orders go to a paper account with simulated fills, so execution quality is precisely what this phase is designed to prove. Backtests behind the expectation band include commissions, spreads and roll costs. Drawdowns are computed on compounded equity and annualised figures use 256 trading days. The methodology is fixed for the duration of the validation; any change would be published as a dated revision.

The expectation band and the evaluation figures start at a dated anchor when a closed model error distorted the track before it; the anchor and its reason are stated on the chart and in the event analysis. The track itself is never restated.

Disclaimer

This is a personal research project, traded with my own capital. Nothing on this site is investment advice, an offer, or a solicitation; no client relationship is created. Figures from the validation phase come from a paper account with simulated execution. Futures trading carries a substantial risk of loss. Content may contain errors and can change without notice. Past performance is not indicative of future results.