Lies, Damned Lies and Mediation Statistics

  • 01 octobre 2012
  • Vance Cooper

While I was not necessarily the strongest student in mathematics through high school, my experience as a mediator has demonstrated the power and persuasion of numbers. There can be no doubt that a series of well planned and sequenced numbers offers a strong form of communication in the course of negotiation. Facility with and expertise of mathematical concepts applicable to personal injury and insurance claims are vitally important when assessing and negotiating claims of this nature.

As a mediator, I formed some empirical thoughts or impressions as to how negotiations proceeded within a mathematical construct. My “seat-of-the-pants” beliefs were:

  • Counsel for a plaintiff or claimant typically demands 2, 3 or 4 times that which they hope or expect to receive by way of settlement;
  • Defence counsel or counsel for an insurer typically offers 1/2, 1/3 or 1/4 of the ultimate amount they expect to pay to achieve settlement; and
  • If the defence opening offer is 10% or more of the plainplaintiff’s opening demand, the case has a probability of settlement.

My opening, at the commencement of mediation, typically incorporates a comment to the effect that I am optimistic that the matter will settle. If I don’t believe the case is capable of resolution, why would the parties or their counsel believe this to be the case? I believed, at that time, that the substantial majority of the matters that came before me for mediation did indeed settle. A comment made by particularly vocal and strident counsel at one of my mediations prompted me to probe deeper into this area.

On one such occasion, this unnamed lawyer quickly interjected and said something to the effect of, “You mediators are all the same. You tell us that 90% or more of your cases settle at mediation. In my experience, I am lucky to settle 50% of my cases at mediation.” I cannot recall if I replied with my inside voice or my outside voice but my thought was, “Perhaps this says more about your style of negotiation at mediation than about the mediation process or mediators with whom you work.” I left the mediation thinking about the exchange and wondering as to whether my seatof-the-pants thought was accurate.

Over the course of approximately 13 months, I recorded data in relation to almost 230 mediations. The substantial majority were full-day mediations but there was a strong collection of half-day mediations. Some of the cases were two-party cases (typically, a plaintiff or claimant and a responding insurer (BI, AB or LTD)). If there were multiple claims (for example, as against BI and AB and / or LTD insurers), this was treated as two or three mediations (since each claim could settle independent of the other). If the plaintiff’s claim was against multiple defendants (for example, a multi-vehicle car accident), this was treated as one mediation.

I recorded the date of mediation, date of loss (if applicable),whether the matter settled or not, opening demand [as made by the plaintiff or claimant], opening offer [as made by the defendant or insurer], final demand (if the matter did not settle), final offer (if the matter did not settle) and the settlement amount (if the matter did settle).

In an appreciable but small percentage of the mediations, there was a “failure to launch”. By this, I mean that in some cases, the content of the brief or something said by counsel before or during the general session of the mediation made it clear to the other counsel that settlement was not going to occur at mediation and, as a result, one party or the other refused to make an opening demand or offer. In some cases, the plaintiff made an opening demand that was met with an opening offer of nil dollars and was told, through me and with permission, that the defence did not see the plaintiff’s claim as having any merit or dollar value for settlement.

I struggled as to whether to incorporate data in relation to these “failure to launch” mediations in my analysis. Some of my mediation colleagues felt that the refusal of one side or the other to engage in a negotiation did not constitute a mediation. Others felt that once the parties and their counsel arrive at and embark on a mediation, it is the mediator’s responsibility to get the parties to engage in a negotiation and if the matter does not settle, this occurs on that mediator’s “watch”. Given that I have had some occasions where I have been able to move one side or the other from an extreme or potentially entrenched position, I felt the appropriate thing to do was to incorporate all of the mediations in the analysis.

In the final analysis, when I looked at all mediations, I found that 73.3% settled. If I excluded the “failure to launch” mediations, then the settlement rate increased beyond 80%.

However, I was much more interested in the relationship between opening demands and opening offers as it might relate to the actual settlement amount and the ability of the parties to achieve settlement. As a result, I engaged the services and assistance of Adam Spinks, who has an educational background in statistics and computer science.

According to Adam’s analysis, cases tend to settle for approximately 30% of the opening demand. This percentage may be somewhat misleading and inflated since the opening demand is typically calculated without regard for pre-judegment interest, costs or disbursements whereas the final settlement amount tends to be recorded as a sum that is inclusive of interest, costs and disbursements. According to Adam, the general rule of 30% does not apply to settlements at very low or very high dollar levels.

When examining settlement values as a function of the opening offer, there is a general trend that settlement is achieved at about 200% to 250% of the opening offer.

I met with him after he completed his analysis and asked him to “dumb it down” for me:

  1. There is a strong relationship between both opening demands and opening offers and the resultant settlement amount.
  2. Settlement demands are more predictive in relation to the settlement amount than opening offers.
  3. When the opening offer is less than 8% of the opening demand, a settlement is very difficult to predict. To my mind, this accords with my belief that when the opening offer is less than 10% of the opening demand, settlement is less likely to occur.
  4. Despite very high opening demands and very low opening offers (approaching a 10:1 ratio), almost 75% of such claims will settle.


Mediation Stats OO, OD and Settlement Values

The opening demand, opening offer, and settlement values for every case (sorted in ascending order for settlement amount). A quick visual analysis suggests that there might be a strong relationship present.

Mediation Stats SS %25 of OO

Above you can see the opening offer (OO) represented as a percentage value of the sttlement (SS) amount. This indicates that, in general, the opening offer is approximately 30% of the settlement amount. However, since the settlement amount increases along the x-axis, it seems that for very low settlements and very high settlements, the general rule of 30% does not hold.

Mediation Stats SS %25 of OD

The same relationship, except between settlements (SS) and opening demand (OD). The results are less consistent than those found for OO. Once again, the trend becomes volatile at either end of the settlement amount spectrum. Aside from that inconsistency, it appears that a general increasing trend can be observed from approximately 200% to 250%. This distribution suggests non-linear relationship between SS and OD.

Vance CooperAbout the Author

Vance Cooper, Laxton Glass LLP founder of Cooper Mediation Inc.