Conventions
USDT throughoutpercentage, where 0.1 means 0.1%charged on both the buy and the sell legslippage, funding, on-chain gas, withdrawal feesrounded to the coin precision; internal maths uses full precisionProfit and price-change maths
The price-change calculator ignores fees, while the profit calculator charges your fee rate on both legs — that is why the same price move shows different returns on the two pages.
investment × (1 − fee rate) ÷ buy pricecoins held × sell price × (1 − fee rate)net proceeds − investmentnet profit ÷ investment × 100%capital × (1 + % change)(end value ÷ initial investment)^(1 ÷ years) − 1Fee maths
Because both legs are charged, annual fees as a share of turnover always equal the fee rate, which makes the absolute cost at different trading frequencies easy to compare.
order size × fee ratefee per trade × 2 × trades per dayfees per day × 365order size × 2 × trades per day × 365Compounding
Compounding is rolled period by period rather than continuously: balance = previous balance × (1 + return per period), with any contribution added before interest is applied.
Frequency and rate must match: "daily" expects a daily rate, "monthly" a monthly rate. The output is a mathematical demonstration and does not imply any asset can sustain a fixed periodic return.
DCA (linear price model)
The model assumes price moves evenly between the start and end values. It explains how averaging works, uses no real historical prices, and must not be read as a return forecast.
start + (end − start) × i ÷ (periods − 1)Σ (contribution ÷ period price)total contributed ÷ coins accumulatedcoins accumulated × end priceDCA backtest assumptions
| Item | Rule |
|---|---|
| Data source | Binance spot daily candles |
| Fetch path | Cloudflare Worker edge proxy with caching |
| Fill price | close of the corresponding trading day |
| End value | coins accumulated × final close in the window |
| Costs | no fees and no slippage |
| Max drawdown | (peak value − trough value) ÷ peak value |
Backtest results depend heavily on the window: the same plan can look strong in one period and weak in another. Compare 3-year and 5-year windows before drawing conclusions.
Break-even and target price
Break-even assumes fees on both legs, so the fee rate enters squared; the target price calculation excludes fees, so the move you actually need is slightly larger.
investment ÷ [coins held × (1 − fee rate)] = buy price ÷ (1 − fee rate)²break-even price ÷ buy price − 1current price × (1 + target profit ÷ investment)Average cost
Average cost is weighted by amount and price, so it differs from simply averaging the entry prices — the gap widens when the buys are unevenly sized.
Σ (amount ÷ price) for every buytotal invested ÷ coins heldcoins held × current price − total investedFX conversion
Rates come from a public feed (open.er-api.com), cached at the Cloudflare edge with a visible update time, and can be replaced by manual rates. Stablecoins are treated as approximately 1:1 with USD, which may introduce a small deviation.
Precision and known limitations
Internal maths runs at full floating-point precision while the interface rounds to coin precision, so hand-checking may differ in the last digit.
Known limitations: slippage and order-book depth are not modelled; funding, gas and withdrawal fees are excluded; backtest fills use closes; price-change and scenario maths ignore corporate-style adjustments. If your real cost structure differs, re-run with your own fee rate.