Average returns should always be used when evaluating stock performance since they are more accurate than geometric returns.

Average returns should not always be used when evaluating stock performance because they do not account for the effects of compounding over time. Geometric returns, also known as compounded annual growth rates (CAGR), provide a more accurate measure of an investment's performance over a period of time, particularly when there are fluctuations in returns.

The geometric return takes into consideration the inflows and outflows of capital, providing a smoothed average that reflects the actual growth of an investment. This is especially important for investments that may experience significant volatility, as averaging simple returns can sometimes present an overly optimistic or misleading picture of performance.

For instance, if an investment has large ups and downs, the simple average may indicate a higher return than what an investor would actually experience if they invested in that stock, due to the impact of loss recovery. In contrast, geometric returns reflect the reality of the investment's performance over time, showing a more precise and realistic picture.

In summary, while average returns can offer some insights, geometric returns are typically considered more accurate for evaluating stock performance due to their ability to account for the effects of compounding and volatility.