The strategy returned +25.2% after an assumed 2 bps bid/offer spread on every trade, against −0.8% for simply holding gold futures. It matched gold through the rallies, then sold, or stood aside from, the falls.
A paper evaluation of live Permutable gold signals, not a marketed product return. Cumulative, unit-notional returns on roll-adjusted gold futures, after an assumed bid/offer spread of 2 bps per trade (half on entry, half on exit); before fees, commission, roll costs and financing.
Gold rose through January, fell sharply from March to July, then recovered. A holder of gold futures finished −0.8%. The Permutable strategy, which reads the direction of gold news every hour and can be long, short or flat, earned 25.2% after trading costs: it joined the rallies and was short or flat through much of the decline. Over the same window the book's own weighted gold P&L, before this page's assumed spread, was +4.3 points.
This page shows the three trades that delivered most of the return, where the outperformance came from, and how the gold signals are read.
The two lines move together through January, when both were long. They separate in March, and again in June, when the strategy was up 16.3% and gold was down 12.5%.
A decision is a confirmed change in the strategy's view of gold. Three trades delivered most of the strategy's return after costs: together they earned 28.1 points, and the other 30 trades netted −2.9. Each trade is recorded in the trade log with its entry and exit dates. The dates and results are the book's own blotter for this window.
Buy confirmed 2 Jan, 16:00. Trade 7–30 Jan.
Hourly news calls turned bullish on central-bank purchases and strong investor demand, with safe-haven support. All six Buy calls in the eight hours up to confirmation cited central-bank purchases.
Gold rose 13.7 points over the same days, and the strategy captured 13.4: it matched gold rather than beating it, and its value was being positioned for the rally.
Sell 20–26 Mar (+2.2%). Buy confirmed 25 Mar, 21:00; trade 26 Mar – 1 Apr (+6.9%).
The strategy exited its long, went short, and flipped to Buy as central-bank purchases and demand returned and geopolitical tension eased.
The two trades earned 9.1% against 4.7% for gold over the same days.
Sell confirmed 28 May, 07:00. Trade 28 May – 10 Jun.
All 23 Sell calls in the eight hours up to confirmation cited India's higher gold import duty and regulatory changes cutting demand; 15 also cited expectations of US interest-rate increases.
Gold moved −8.1 points over the same days, so the short trade earned 7.8% where a holder lost. June closed with the strategy up 16.3% and gold down 12.5%.
| Trade | Direction | Dates | Strategy | Gold, same days | Versus gold |
|---|---|---|---|---|---|
| January rally | Buy | 7 – 30 Jan | +13.4% | +13.7% | −0.3 |
| Break lower in May | Sell | 28 May – 10 Jun | +7.8% | −8.1% | +15.9 |
| Turn in March | Buy | 26 Mar – 1 Apr | +6.9% | +7.1% | −0.2 |
| September recovery | Buy | 24 – 29 Sep | +5.7% | +5.6% | +0.1 |
| June continuation | Sell | 22 – 25 Jun | +4.5% | −4.3% | +8.8 |
Gold's move is the change in cumulative long-only return between each trade's open and close; “versus gold” is the strategy's return minus that move, in points. A long trade can at best match gold over its own days, so the long trades sit near zero. Trade returns are unit-notional and after the assumed 2 bps spread; a long trade nets slightly below gold's move for that reason.
The long trades matched gold, as any long trade does: across the window the 18 long trades earned 16.3% after costs (16.7% before) against 16.7% for gold over the same days. The outperformance came from two other places. The 15 short trades earned 8.9% after costs (9.2% before) while gold moved −9.7 points over those days, a gain of 18.9 points on holding gold before costs. And on the days the strategy had no position, gold moved −7.7 points, which the strategy neither made nor lost. That is +7.7 points versus holding gold. Trading costs took 0.7 points: 33 round trips at 2 bps each. The book's own weighted gold P&L over this window is +4.3 points, before the spread assumed here.
Points versus holding gold, 1 Jan – 28 Oct 2026, unit notional. The first three rows are before costs; the total is after the assumed 2 bps spread. Components are unrounded, so they may differ from the headline figures by 0.1. Across 215 weekdays the strategy was long on about 60% and short on about 20%.
Permutable's gold signal reads gold news and market developments each hour and returns Buy, Sell or Neutral. A change of view is acted on only after consecutive hourly readings agree. That confirmed view sets the position for the day: long, short or flat. The rules were fixed on data to 31 December 2025.
The figures cover 1 January to 28 October 2026. They are a paper evaluation of the same gold trades as the live book, compared with holding the same gold futures. Returns are after an assumed 2 bps bid/offer spread on each trade, and before fees, commission and financing. The book's weighted contribution quoted above is taken from the live book's equity file and does not deduct that spread.
Past performance is not a guide to future results. This page is for information only and is not investment advice or a recommendation to buy, sell or allocate capital.
Discuss the dataset or API. Email enquiries@permutable.ai or book a demo. See the live commodity book on the Performance page.
Get in touch