The article explores how accurately estimating systematic risk (beta) is crucial for financial applications. It shows that there is a type of risk related to the day when data is collected for shares on the Johannesburg All Share Index. To improve beta estimation, a new method called volume-weighted-average-price (VWAP) is suggested for use when considering this reference-day risk. The study also tests the risk-reward relationship in the capital asset pricing model (CAPM) using a graphical time-series approach. The findings indicate that the CAPM does not perfectly capture the risk-reward trade-off when using beta as a measure of systematic risk.