Accuracy, with the sample sizes attached
Per-strategy trailing windows computed from closed-trade records — not marketing claims. Every figure carries its n, losing windows are published identically to winning ones, and the methodology below explains exactly how each number is computed — including what these figures are not: paper-trading (simulated) and live-money results are currently reported together.
Methodology first, numbers later. This page publishes how figures will be measured before any track record exists. Collection dates appear here only once the snapshot API reports them — we do not claim data is accumulating unless it says so. We will not backfill, cherry-pick a start date, or publish a window before it has real data behind it. If a section below is empty, that is why.
Important: Tradewink is not a registered investment adviser or broker-dealer. Figures on this page are platform-reported and not independently audited. Past performance does not guarantee future results. Trading involves substantial risk of loss.
Hypothetical performance disclosure: The closed-trade records behind this page do not separate paper-trading (simulated) executions from live-money executions, and most platform trading runs in paper mode. Treat every figure here as including simulated results — hypothetical performance with the limitations described in the disclaimer at the bottom of this page. Simulated results do not represent actual trading.
Loading published windows · Snapshots are written once per trading day after market close (about 4:20 pm ET) · Served from a cache refreshed at most hourly
Methodology
Closed trades, not funnel telemetry. Figures come from closed-trade analytics records — the tables our own risk systems reconcile against — not from signal counts, intents, or lifecycle telemetry. If a trade did not close, it is not in these numbers.
Paper and live trades are currently mixed. The closed-trade table does not yet carry a per-row paper/live flag, so every window on this page combines paper-trading (simulated) executions with live-money executions — and most platform trading runs in paper mode. Until a per-row split ships, all figures here are hypothetical performance. Where available, the paper-vs-live mix of submitted executions is disclosed in the freshness line above, sourced from broker submit records.
Recovered positions excluded. Positions discovered at a broker without recorded entry context ("recovered" rows) carry placeholder data and are excluded from every window — filtered when snapshots are written and re-filtered when this page's API reads them. Synthetic benchmark-agent rows are excluded the same way. Paper-trading executions are not excluded; they are part of the figures, per the disclosure above.
Snapshot cadence. A snapshot job writes one row per strategy per trading day, shortly after market close (about 4:20 pm ET). This page reads those daily snapshots through a cache refreshed at most hourly — numbers here can lag reality by up to a day by design, which also means they cannot be quietly rewritten intraday.
Definitions. Win rate is wins ÷ closed trades, where a win is a closed trade with positive P&L percent. Expectancy is (win rate × average win) − (loss rate × average loss), in percentage points per trade. Profit factor is gross winning P&L ÷ gross losing P&L. Max drawdown is the largest peak-to-trough drop of the cumulative closed-trade P&L-percent curve within the window, in percentage points.
n is disclosed, always. Every figure states the number of closed trades behind it. A win rate over 6 trades is noise; over 200 it starts to mean something. We publish the n so you can make that judgment — and small-n windows are published anyway rather than hidden.
Losing periods are shown. Windows are published on a fixed daily schedule regardless of outcome. A losing 30-day window renders in the same table, same format, same styling as a winning one.
External verifiability: planned, not yet live
Today these figures are platform-reported: you are trusting that our snapshot pipeline works as described above. We are committed to shipping external verifiability — publishing a cryptographic hash and timestamp of each signal before its outcome is known, so third parties can confirm records were never edited after the fact. Until that ships, we will not claim these numbers are independently verifiable, and you should weigh them accordingly.
How to read a track record (please read)
Short windows reflect the market regime they were measured in and may not persist. Win rate alone is meaningless without expectancy and profit factor — a high win rate with large losers still loses money. Aggregate results do not predict your results: strategy selection, broker, account size, and risk settings all change outcomes. Nothing on this page is investment advice.
Related pages
For live platform-wide stats see Performance and Live stats. For how the system works, see Architecture.
Tradewink is not a registered investment adviser, broker-dealer, or financial planner. All data, signals, and analytics on this page are for informational purposes only and do not constitute investment advice, financial advice, or a recommendation to buy or sell any security.
Past performance does not guarantee future results. Trading involves substantial risk of loss, including the possibility of losing more than your initial investment. You are solely responsible for your own trading decisions.
Hypothetical or backtested performance results have inherent limitations. Unlike actual trading records, simulated results do not represent real trading and may not account for the impact of market liquidity, slippage, or all transaction costs. No representation is made that any account will or is likely to achieve profits or losses similar to those shown.