Cumulative abnormal returns (CAR) computed around acquisition announcements are widely considered market-based assessments of expected value creation. We show that announcement returns do not correlate with commonly used and new measures of ex-post acquisition outcomes. A simple characteristics model using standard information known at announcement can predict outcomes reasonably well, and CAR fails even to capture the prediction from this model. We present evidence suggesting that information about the standalone acquirer dominates CAR and, therefore, makes it virtually impossible to extract deal-related information. We conclude that CAR is an unreliable measure of expected value creation.