10,000 held in stablecoin
A fictional 10,000 deposited on January 1, 2026 and left entirely in stablecoin. No asset picked, no strategy applied, no trade. This is what it has done since — and it is the reference every other figure on this site is measured against. Turn on a comparison below to see what the same amount would have done otherwise.
30 intermediate stopping points what the chain would be worth if it had stopped there
The stablecoin line always stays on: it is the reference, drawn in neutral ink. Every comparison starts on January 8, the day the first proposal was applied, marked by the vertical line — a composition has no value before it exists. A stopping point applies every proposal up to that date and then holds: a reallocation starts from the portfolio the previous one produced, so none can be taken in isolation.
2 decisions, each applied in full
Each proposal was produced from one week of data and nothing else — the daily candles of the seven days before, and the news published in that same window. Every proposal is then applied in full, so the next one starts from the portfolio the previous one produced. The chain is now worth 8,903 (−11.0%).
January 8, 2026 · portfolio 10,000 · 55% invested
| Asset | Share | Amount |
|---|---|---|
| USDCUSDT | 45.0% | 4,500 |
| BTCUSDT | 20.0% | 2,000 |
| ETHUSDT | 15.0% | 1,500 |
| SOLUSDT | 10.0% | 1,000 |
| XRPUSDT | 10.0% | 1,000 |
January 14, 2026 · portfolio 10,271 · 55% invested · fees 2.01
| Asset | Share | Amount | Change |
|---|---|---|---|
| USDCUSDT | 45.0% | 4,622 | — |
| BTCUSDT | 22.0% | 2,260 | +2.0 pt |
| ETHUSDT | 10.0% | 1,027 | −5.0 pt |
| SOLUSDT | 10.0% | 1,027 | — |
| BNBUSDT | 7.0% | 719 | +7.0 pt |
| XRPUSDT | 6.0% | 616 | −4.0 pt |
This is a retrospective test, and it proves nothing. Each proposal was given data cut off the evening before, but the model was trained after this whole period and may know what followed — a curve that beats Bitcoin here is not evidence of judgement. Transaction fees of 0.1% per order are deducted from the second decision onwards (2.01 so far); the initial deployment out of stablecoin is not charged, which flatters the chain by roughly 11.
What this line is, and is not
It is not a straight line drawn for convenience. These are the real daily closes of USDCUSDT, one per day, read from the same cache every other page uses. A stablecoin is not perfectly stable: over this window the value moved between 9,984 and 10,009 — 0.25% between its lowest and highest point, invisible at this scale. A missing close leaves a gap; no point is ever invented.
The vertical scale never tightens onto it. Bounds are rounded outwards to the nearest thousand and kept at least 2,000 wide, on first paint and while you zoom. A scale fitted to those few dollars would turn 0.25% into a mountain.
It is the hardest reference to beat in a falling market. Doing nothing costs nothing and earns nothing — so any strategy ending below this line has, on this window, destroyed value rather than created it. See what the site's own recommendations did against it on the track record page.
Simulated on past data. No order was ever placed and no account exists. This is not the track record of a managed account, and nothing here is an investment recommendation.