Given their prevalence in recent years, earnings management and financial restatements have been at the center of much of the discussion surrounding corporate malfeasance. This study builds a probability model for predicting the likelihood of earnings restatements by analyzing the trends in and the deviations from the industry averages of the return on assets, accounts receivable turnover, net profit margin, and cash flow to net income measures. Data are obtained for a sample of 104 firms (restating as well as non-restating) for the 2000 to 2001 period. The results suggest that the deviation from the industry average of the accounts receivable turnover and the variability in the cash flow to net income are good barometers for detecting fraudulent accounting. Potential restating firms have higher accounts receivable turnover rates than their industry counterparts and downward trends in their cash flow to net income, signaling the likelihood of a restatement, at least in the current study.