Cortik Alpha · live testnet results
loading…⚡ Live activity
🐢 Slow lane today
Test wallets
How Cortik Alpha works
Once a day, after the 00:00 UTC close, it looks at whether Bitcoin and Ethereum have been rising over the last 20, 60 and 120 days. In an uptrend it buys; when the trend fades it steps aside. Position size follows the market's calmness: smaller when prices swing wildly, larger when they're calm. It changes positions only when the change is worth it, roughly once a week.
Every hour it scans the most-traded coins with trend, breakout and mean-reversion strategies, and trades only signals that pass strict quality checks. It's shown here for transparency: in our tests it did not yet prove a lasting edge after costs, so it stays an experiment on testnet.
Every trade starts with a stop-loss on the exchange. Some test wallets also try smarter exits: LockProfit locks in part of a gain as a trade moves up, and the trailing stop follows the price at a safe distance. These are being compared live.
Defensive, balanced and aggressive differ in how much each trade may risk, how many positions can be open, and the target level of price swings. Higher risk means bigger possible gains and bigger possible drops; no mode can avoid losses.
How we test before anything goes live
- Every idea is tested on 5 to 7 years of real market history, with trading fees, slippage and hourly funding counted.
- Tests run month by month without seeing the future, and as one real account holding several coins from one balance.
- An idea only goes live if it's clearly better, including a luck check (at least 90% chance the improvement is real, and still better over the last 24 months) and it helps in at least two of three risk modes.
- So far most ideas were rejected. The one that passed every check is the slow lane: daily trend-following on Bitcoin and Ethereum with volatility-based sizing.
| Slow lane backtest, May 2020 – Aug 2026 | Defensive | Balanced | Aggressive |
|---|---|---|---|
| Total result | +40% | +118% | +185% |
| Worst drop from a high | 6% | 13% | 21% |
| Holding BTC+ETH instead (no costs) | +1,007%, but a 76% worst drop (2022: −65%) | ||
Caveats: a backtest is not a forecast. The slow lane trades only two coins, so its results rest on a limited number of big trends, and much of the gain came in 2020–21. It still made money in every version of its settings we tried and from every later starting year, which is why it runs live.
Safety
- Every position has a stop-loss resting on the exchange itself, so it's protected even if our servers stop.
- Trading keys can trade only: they can never withdraw funds.
- A kill switch stops all new trades at once; exposure is capped per risk mode.
- Testnet prices are checked against the real market; coins that drift more than 1% are not traded, so these results stay meaningful.
Data refreshes every minute.