When a company reports earnings, any seasoned analyst worth their salt can break down revenues and expenses by business line and geography, plot historical trends, adjust profitability based on alternative assumptions, and so forth, probably with one hand tied behind her back. Surely some do it better than others in some normative sense, but there’s really no right or wrong when the task at hand is to describe what happened. But if you ask the same analyst to then forecast the company’s next quarter profits, that’s when she will not only use both hands but they are shaking from stress because, in forecasting, she can definitely be wrong. The former exercise is deterministic, whereas the latter demands to know “so what?”
The same great chasm between attribution and prediction exists when analyzing managers. Attribution analysis breaks down exposures to factors and geographies, plots historical trends, and provides a bunch of metrics. Barring dirty data or software bugs, there’s no right or wrong in attribution analysis. It can tell investors things like the manager had a Sharpe of 0.6, a momentum beta of 0.3, and exited positions too early by an average of 5 days. That feels interesting, but so what?