Ohlson O-Score
EliteNine-input bankruptcy probability — a Z-Score complementThe Ohlson (1980) O-Score is a nine-input logit model of bankruptcy probability, computed from the latest fiscal year. Asset size uses dollars in millions in place of the original GNP deflator (a constant rescaling that keeps the ranking intact). Inputs are winsorised: because the model uses book equity, heavy-buyback firms with negative book equity would otherwise read as high-risk even when healthy, a known divergence from the market-equity Altman Z, so cross-check both. Not investment advice.
How it works
James Ohlson's answer to Altman. Where the Z-Score weights five ratios into a number you then have to interpret, the O-Score runs a logistic regression over nine inputs and hands back a probability.
It leans harder on liquidity and on direction of travel: whether current liabilities exceed current assets, whether the company lost money in the last two years, whether earnings are improving or sliding. Read it alongside the Z-Score rather than instead of it. They disagree often, and the disagreement is the interesting part.
When it misleads
It has no idea what a bank is. Financial companies carry leverage as a matter of business model and score as though they are about to fail.
A probability is not a forecast about this company. Thirty percent means companies whose books looked like this failed at roughly that rate in Ohlson's sample.
This screen ranks the most fragile names first, so we don’t publish its list — naming companies as distressed on a public page isn’t something a screening model should do on its own. What today’s run found, in aggregate: