4 PerspectivesApril 22, 2010

Decision Trees in Mediation: A Few Examples

It’s no secret that I believe decision trees can make a difference as you try to settle your next lawsuit, and my series on decision trees will tell you why.  But I’m not the only one.  Your comments to my prior posts, our follow-on discussions since then, and a little research confirm that a confident minority of mediators and litigators use them, too.  This post is the first of three over the next few weeks that will give you real-life examples of how decision trees are used to settle disputes.

How do mediators and advocates use decision trees in mediation? A month or two ago we had a great discussion among the Commercial and Industry Arbitration and Mediation Group on LinkedIn styled “Do You Use Decision Trees in Your Mediation Practice?” More than a few mediators and negotiators spoke up, with each providing insight into how decision trees can help get your case settled.  I’ll highlight some of those tips, and a few more from other sources, in this and subsequent posts over the next two weeks.

A Better Way to See the Dispute

Portland area mediator Debra Healy summarized the thoughts of many when she said that mediators can use decision trees as a “visual, tangible reality check”:

It can be so difficult for a client with no experience with litigation to even fathom the scope of uncertainties involved. Add to this that the client may not understand when his/her attorney is posturing. If his/her attorney says something, the client takes it at face value and hangs on every word. . . . I primarily use decision trees as a visual, tangible reality check.

A Way to Work with “Quantitative Sorts”

Pittsburgh mediator and civil engineer Rebecca Bowman mediates complex technical disputes, and provides some insight into why decision trees work well with “quantitative sorts”:

Engineers and many accountants generally prefer finite things. It can be extremely powerful to have a clean, concrete decision tree to evaluate risk. Quantitative sorts find it very comfortable to wrap their heads around a 60% probability of an outcome of X dollars. Fold in a time prediction and they can fold in a time-value-of-money factor. If you can show the parties a way to reduce the mushy, often gut-feel outcome predictions to something concrete that they can evaluate, it has been my experience that they are much more likely to march directly to a resolution. Of course, this requires that you understand the decision tree intimately so that you can help them to make appropriate predictions. But that’s what they pay us the big bucks for, right?

A “Dash of Ice Water” for Clients

Dallas mediator Richard Faulkner described how decision trees can serve as “a dash of ice water” to clients’ expectations:

If the parties, or any of them, are inclined toward using logical approaches, [decision trees] can be very effective. However, I use it only after I have obtained as much of the decision point percentage analysis information as possible from the party’s attorney. That allows me to use a combined set of techniques. First, I ask counsel to provide me an evaluation of the probable outcome if the case is not settled. That is usually offered in generalized language like, “a good chance”, most probable etc. We all know as experienced lawyers that means “50 -50″, “60 – 40″ etc to us. The client is typically thinking far higher percentages. I then ask for the lawyer to quantify in numbers their percentage evaluations or evaluation ranges for me. Those “numbers” and “ranges” tend to surprise many clients, which then helps me change the client’s expectations through rigorous reality testing. After that dash of ice water, I will sketch out the “Litigation Risk” analytical decision tree based on their lawyers numerical percentages. We then work towards settlement using that information and modifying it as necessary when new information is learned. It can be a very powerful technique with any parties that are mathematically inclined or impressed. Of course, I always reminded my advanced mediation students to remember Mark Twain’s quip, “There are lies, there are damnable lies, and then there are statistics.” . . . If this does not work in the mediation, it can then later be used as the basis for a “Mediator’s Dynamite Charge” follow up.

A Credible Perspective for Upper Management

New York litigator and arbitrator Philip J. Loree Jr., who blogs at the Loree Reinsurance and Arbitration Law Forum, uses decision trees as he analyzes cases with his clients’ executives:

I have found decision trees to be a valuable tool to help advise clients of the settlement value of complex cases. Often times the client is looking for something with some basis in logic and science to help convince upper management to authorize a settlement number, and a decision tree can provide that kind of support. Predicting settlement value is part art and part science and one should draw on both disciplines when putting together a decision tree. The nice thing, though, is the end product looks more like science than art, and that can boost its credibility in the eyes of upper management.

Helping Clients See the Hurdles

UK mediator Philip Hesketh argues that clients see the case — and understand the rigor of their counsel’s analysis — through decision trees:

Decision trees can help clients who want a valuation based on more than your legal intuition. Working through a decision tree with your client will show them that you have vigorously evaluated each aspect of their claim. They will know each and every hurdle in their way and understand how you have assessed each.

(Hat tip to Jacob Ruytenbeek at the Decision Tree Analysis Blog for leading me to Philip’s article.)

Settlements that Stick

Washington lawyer Patrick F. Hofer, a former student of decision tree pioneer Marc Victor, commented on a prior Settlement Perspectives post that decision trees help parties get to the actual value of the case, resulting in settlements that stick:

I have successfully used decision trees to settle dozens of cases . . ..  I have used them to convince opponents of the reasonableness of my client’s deal – from in-house counsel at a large multi-national conglomerate, who viewed themselves as God’s gift to the law, to the owner of a dry-cleaner who didn’t have a high school diploma. In every case I settled using a decision tree, the deal stuck and did not come unglued or get retraded, because both sides saw the value in the deal they made. Decision trees are amazingly effective, efficient and powerful tools to get parties to settle their differences.

You Need More Than a Tool

There’s no question that decision trees can improve your mediation results, and the perspectives above show us why.  But you need more than a tool — ADR Toolbox’s Don Philbin has reminded us that decision trees are helpful, but at the end of the day they’re “just another tool in the mediator’s toolbox”.

So what does get the deal done?  Bethesda mediator Daniel Preston Dozier helps us with that:

[A decision tree] can look more like science than art; the ‘science’ can assist a negotiator sell a settlement to internal decision-makers. But in the end the settlement is generally art — perhaps realistic art, but art none the less.

Yes, settlement is art — but add a little science to your art with a decision tree next time.  You’ll be glad you did.

[UPDATE:  For for more on Decision Tree Analysis, Settlement Perspectives' series on decision trees includes:

1. Decision Tree Analysis in Litigation:  The Basics;
   
2. Why Should You Try a Decision Tree in Your Next Dispute?;
3. Advanced Decision Tree Analysis in Litigation:  An Interview with Marc Victor, Part I;
   
4. Advanced Decision Tree Analysis in Litigation:  An Interview With Marc Victor, Part II;
   
5. Decision Trees in Mediation:  A Few Examples (this post); and
6. Avoiding the Limitations of Decision Trees:  A Few Tips from Mediators Who Use Them.]

Categories: Decision Trees,Mediation,Negotiation,Settlement,Tactics

I welcome comments on this or any post. Please feel free to comment in the “Add Your Own Perspective” area below, whether you use your actual name or pseudonym. If you don’t feel comfortable commenting directly on this site, I can easily be reached by email at jd@johndegroote.com.