The The New York Times reported last week (8 Aug 2008) on a study analyzing the cost of not settling law suits. Defense lawyers make bad calls not to settle in almost 1 out of 4 cases. Can lawyers improve that record? 

Study Finds Settling Is Better Than Going to Trial reports that that plaintiffs would have been better off 61% of the time accepting settlement offers; defendants would have been better off 24% of the time. The average cost of defendants’ bad calls was $1.1 million.

Is this just the cost of doing business and practicing law? The Times also reports

As part of the study, which is the biggest of its kind to date, the authors surveyed trial outcomes over 40 years until 2004. They found that over time, poor decisions to go to trial have actually become more frequent.
‘It’s peculiar if any field is not improving its performance over a 40-year period,” Mr. Kiser [a study author] said. ‘That’s a troubling finding.’

Two ways come to mind to improve outcomes. One is consistent use of litigation risk analysis. I have blogged about decision trees, a technique to lay out the legal and factual issues, key decisions, and probabilities to determine an expected case value. It’s hard work and involves judgment but forces lawyers and clients to be clear about a case and its odds. Another way is predictive markets.

Whether either way would help is an empirical question. The typical lawyer’s attitude, however, works against systematically assessing how to improve outcomes. The article reports that :

Several lawyers were dismissive of the study, noting that the statistics mean nothing when contemplating a particular case, with its specific facts and legal issues, before a specific judge. They stressed the importance of a lawyer’s experience.

In other words, “trust me, I’m the expert.” Experience is indeed critical but why not improve upon it if, in fact, analytic techniques actually do improve outcomes? The presumption today is that litigators need not do risk analysis. That 1 of 4 cases mistakenly go to trial and the record has not improved in 40 years, however, shifts the burden of proof onto expert litigators to show why they should not conduct risk analysis.

I fear that for some lawyers, the right adage is less “trust me, I’m the expert” and more “what, me worry?” (with thanks to Alfred E. Neuman).