**Use adjusted close for a simple return-based backtest; use actual prices plus explicit corporate-action accounting for a realistic trading simulation.** They answer different questions.

Regular close tells you the price quoted that day. Adjusted close makes prices comparable across splits and distributions. Its historical values are synthetic, but that does not make them less useful.

For example:

- **A 2-for-1 split:** A stock closes at $100 before the split and $50 afterward. You now own twice as many shares, so you have not lost 50%. Comparing raw closes alone would incorrectly report that loss.
- **A $1 cash dividend:** Suppose a stock closes at $100 before going ex-dividend and $99 afterward. An eligible shareholder has $99 of stock plus a $1 dividend entitlement, so the price decline alone does not represent their total return.

The practical choice depends on what you are testing:

| Backtest purpose | Appropriate data |
|---|---|
| Buy-and-hold performance or a strategy based on percentage returns | Split-and-dividend-adjusted close, after checking the provider’s methodology |
| Indicators intended to ignore mechanical corporate-action price changes | Consistently adjusted prices |
| Order fills, share quantities, cash balances, and transaction costs | Actual prices, with splits and dividends handled explicitly |
| Rules involving absolute prices, such as “buy stocks below $10” | Prices available at that historical date |

Backtesting systems explicitly distinguish these approaches; for example, QuantConnect offers raw, split-adjusted, dividend-adjusted, and total-return modes. [Data normalization documentation](https://www.quantconnect.com/docs/v2/writing-algorithms/securities/asset-classes/us-equity/requesting-data)

Keep three pitfalls in mind:

1. **Do not count dividends twice.** If your return series already incorporates dividends, adding dividend cash again overstates performance. [Corporate-action documentation](https://www.quantconnect.com/docs/v2/writing-algorithms/securities/asset-classes/us-equity/corporate-actions)
2. **Keep price fields consistent.** Comparing an adjusted close with an unadjusted open, high, or low can produce misleading signals.
3. **Watch absolute-price rules.** Today’s adjusted history can incorporate later corporate actions, changing whether an old price appears below a threshold. Use point-in-time data for those rules. [Look-ahead bias documentation](https://github.com/QuantConnect/Documentation/blob/master/03%20Writing%20Algorithms/01%20Key%20Concepts/10%20Research%20Guide/07%20Look-ahead%20Bias.html)

For your initial backtests, **adjusted close is usually the useful starting point if you are calculating returns**. As you add trade execution and portfolio accounting, use actual prices and model corporate actions explicitly.