Not all stock recommendation changes are equal. In a sample constructed to minimize the impact of confounding news, relatively few analyst recommendation changes are influential in the sense that they impact investors' beliefs about a firm in a way that could be noticed in that firm's stock returns. More than one-third of the stock-price reactions to analyst recommendation changes have the wrong sign and only approximately 10% have significant stock-price reactions at the 5% level using an extended market model. We find that the probability of an influential recommendation is higher for leader analysts, star analysts, away-from-consensus revisions, revisions issued contemporaneously with earnings forecasts, analysts with greater relative experience, and those with more accurate earnings estimates. Growth firms, small firms, high institutional ownership firms, and high prior turnover firms are also more likely to have influential stock recommendations. Strikingly, analyst recommendations are more likely to be influential after Reg FD and the settlement. Finally, influential recommendations are associated with increases in stock volatility and large absolute changes in consensus earnings forecasts.