These are the fixed inputs of the owner-approved EUR 10k compounding research candidate. They describe a measured scenario, not a promise about live execution and not a completed USDT test.
Candidate boundary
What was actually held constant
Strategy version 2.0.0 on Freqtrade 2025.11.2, data through .
Market
Binance Spot
Pairs
BTC/EUR + ETH/EUR
Quote asset
EUR
Candle timeframe
4h
Initial wallet
10,000 EUR
Stake mode
Current wallet (unlimited)
Tradable balance
99%
Maximum open trades
1
Compounding
Enabled
Fee per fill
10 basis points
Base slippage model
none-v1
Missing-candle fill
Disabled
Trailing
OFF
Result truth model
Three terminal values, not one convenient total.
End-of-period bookkeeping must not turn an unfinished position into a successful strategy exit.
01
Realized equity
Only completed strategy trades contribute realized profit or loss.
02
Mark-to-market equity
Open positions use the last complete candle close and stay labelled open at end.
03
Engine terminal equity
Freqtrade's terminal value is retained for reconciliation, not silently promoted to the headline.
Important limits
What this scenario does not establish
No live parity yet. The intended production settings have not yet been proven byte-for-byte against this compounding research setup.
No repaired market rows. The reviewed source gap is kept visible; the loader runs with missing-candle filling disabled.
No fill guarantee. Candle-based limit-order simulation, fixed fees and deterministic stress cannot reproduce queue position, latency or every live fill.
No pair-only decomposition shortcut. BTC-only and ETH-only scenarios rerun from fresh wallets; they are not slices of the combined wallet.
No terminal force-exit claim. Engine bookkeeping at the boundary is normalized as open at end.