We reconcile the empirically flat relation between historical betas and stock returns (flat security market line) with the common usage of the CAPM based on historical betas in valuation. Analysts bias cash flow growth expectations upwards for high-beta firms, so that the value-reducing effect of higher historical systematic risk cancels out and buy/sell-recommendations remain unrelated to beta. The association between beta and growth overestimation is driven by estimates conventionally used in the industry (e.g., Bloomberg betas), suggesting that analysts adjust growth expectations to offset beta's valuation effects, instead of exhibiting a coincidentally higher overoptimism for higher-beta firms.