« Help rank the most publicized recent practices in law departments | Main | How to increase the “deep smarts” of inside lawyers »
Estimating the probabilities of a legal event occurring
Clients often plead with their in-house counselors to estimate the likelihood of, for instance, a patent holding up to an infringement claim or an indemnity provision being triggered. Researchers at Wharton’s Risk Management and Decision Process Center have demonstrated that a lawyer (actually, they generalized to all people) has the best change of estimating expected value when a variety of low-probability events are aggregated to generate a probability. For example, “estimate the probability that there will be either a reversal, a remititure or an en banc decision” as compared to estimating the likelihood of each possibility individually.
Other studies suggest that lawyers are more effective when assessing possible outcomes relative to low probability events they are familiar with, such as “estimate the risk of a $10 million plus verdict versus the risk of having a traffic accident.”
Finally, research shows that decision makers evaluating low probability, high impact events tend to either over-insure, assuming the occurrence of the event was inevitable, or ignore the event entirely, thinking “that can’t happen to me.” Lawyers, tending to conservativism, are probably more likely to over-react to the low probability, high impact event (the European Union rejecting the merger, the class action being certified, the 8K disclosure clobbering the share price). The source of these points was The Professional Services Firm Bible, John Baschab and John Piot, Eds. at pgs 32-33.
Posted on July 15, 2005 at 08:54 AM in Thinking | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d834519fb069e200d8351c697853ef
Listed below are links to weblogs that reference Estimating the probabilities of a legal event occurring:
Comments
You can follow this conversation by subscribing to the comment feed for this post.

