Import Settings That Change Your Numbers
Most "my metrics look wrong" moments trace back to four import settings. None of them affect your raw trade P&L — but they change every percentage, every score input, and every time-based chart computed from it. This short lesson makes sure you get them right the first time.
Warning
If a freshly imported strategy shows a suspicious CAGR, drawdown percentage, or dollar P&L, check these four settings before doubting the strategy. They're the usual suspects, in this order.
1. Initial Capital
The starting account balance for percentage math. A $10,000 profit is a 100% return on $10K starting capital and a 10% return on $100K — same trades, wildly different CAGR, drawdown %, and every chart with a % toggle.
Set it to: the capital you'd realistically allocate to this strategy, or the account size used in the backtest. Consistency matters more than precision — if you compare strategies, give them comparable capital assumptions.
2. Slippage & Commission
Trading costs turn many "profitable" backtests into break-even reality. The import wizard offers three modes:
| Option in the wizard | What it does | Use when |
|---|---|---|
| No Slipp & Comm | Imports P&L exactly as-is | Your export already includes costs |
| Add Strategy Slipp & Comm | Applies the costs recorded in your strategy file | Your platform exported per-trade costs |
| Add Database Slipp & Comm | Applies the stored costs from your Instruments catalog | You maintain realistic per-contract costs centrally (recommended for futures) |
Tip
The most common double-counting mistake: your platform already subtracted commissions AND you apply database costs on top. If your average trade looks worse in AlgoChef than in your platform, check this first.
3. Instrument (and point value)
For futures, the instrument selection tells AlgoChef the point value — how many dollars one point of price movement is worth — plus margin and cost defaults from your Instruments catalog. If your file records P&L in points rather than dollars, the wrong instrument means every dollar figure in the report is scaled wrong.
The wizard's instrument type options are Futures, Stocks / ETFs, Forex, and Crypto. For futures, pick the exact contract (ES, NQ, GC, …) — micro vs full-size matters ($5/point vs $50/point on the S&P).
4. Strategy Timezone (and Time Format)
The timezone anchors your trades to calendar days and market sessions. Get it wrong and trades slide across day boundaries — session analysis, day-of-week stats, and daily aggregation all shift. Pick the timezone your platform recorded timestamps in (often the exchange's timezone, not yours).
Time Format only needs selecting when your file has entry/exit time columns mapped — for date-only files the wizard won't ask.
Quick sanity check after import
Thirty seconds that catch 90% of setting mistakes. Open the new strategy's Quality Report and check:
- Net Profit matches what your platform reported (roughly — cost settings explain small gaps)
- Total trades matches your export's row count
- First and last trade dates match your backtest window
- Average trade isn't suspiciously better or worse than your platform showed (cost double-counting shows up here)
If something's off, delete the strategy and re-import with corrected settings — imports take seconds, and there's no penalty for redoing one.
Info
Settings like Initial Capital live with the strategy, not the file. Importing the same file twice with different settings produces genuinely different reports — that's by design, not a bug.
Next lesson
Your data is in and the numbers are trustworthy. Now read the whole report properly: Understanding Your Dashboard →