LSF Index Brochure
The Litigation Settlement Failure Index (LSF Index) is the world’s first legal metric to serve as a quantitative early-warning indicator of trial-bound disputes.
Revised November 20, 2017
The Economic Fingerprint of Settlement Failure
Not all legal disputes are created equal. The subset of lawsuits that proceed to trial and adjudication are characteristically different from those that are resolved through settlement.
In order to reach a courtroom and obtain a judgment, disputes must first survive the settlement bargaining process and resist the economic forces that might ordinarily encourage or compel a negotiated solution.
Having been filtered along economic lines, it should come as no surprise that the disputes that fail to settle have a tell-tale economic fingerprint.
The Litigation Settlement Failure Index is that fingerprint.
Settlement Failure Risk Analysis
The majority of legal disputes end in settlement. So when settlement bargaining fails and cases proceed to a court or arbitral hearing, it can often come as a surprise, and one that is not always welcome. A number of factors can give rise to a permanent bargaining impasse, and the reasons for a particular settlement failure are not always immediately obvious. To say that the parties disagreed is oversimplifying. We know they disagreed – but why? If we can answer this question, we can begin to anticipate when disputes will fail to settle.
Fortunately, help is at hand. Research shows that there are signs of impending settlement failure embedded in the economic landscape of trial-prone disputes – buried in the mix of trial expectations, costs and uncertainties of legal conflict. In fact, theory allows us to draw an explicit mathematical connection between the ‘economic state’ of any legal dispute and the likelihood of it being settled – or not.
The Litigation Settlement Failure Index™ (‘LSF Index™’) is the result of such an analysis. Based on an advanced proprietary model of litigation and settlement decision making, the Index extracts the implied theoretical probability of settlement failure from a detailed economic description of any lawsuit.
The science of predicting trial or settlement.
Forewarned is Forearmed
Using the LSF Index and related analytics, litigants and counsel can now be alerted as to problem legal claims and begin to anticipate when settlement failure becomes probable or even inevitable.
Forewarned being forearmed, the Index can be used to head off what may otherwise be an impending and certain trial or arbitration. Understanding the true extent of settlement failure risk and the economic factors contributing to it, litigants can take proactive steps to avoid the problem – an analytic process we call Settlement Risk Engineering™.
In addition, law firms can use the LSF Index as part of their case intake process and early-stage case assessment. Knowing whether a dispute is more settlement- or trial-prone can allow firms to optimize case selection and allocate legal resources more efficiently.
By applying the Index across claim portfolios, legal cases can now be ranked according to their likelihood of ending up in a courtroom. For large law firms and institutional litigants, the LSF Index becomes an invaluable portfolio management tool.
The essential elements of contingent disputes being the same, there is reason to believe that the LSF Index also has application to disputed insurance claims.
A Focus on Failure
By definition, the LSF Index frames the question of settlement in terms of a failure risk analysis. The reason for this is simple: because as practitioners and managers of risk, we care most about the causes and likelihood of rare and costly events. And it is the rare failure of litigation settlement bargaining that gives rise to costly trials.
This failure-centric approach to analysis finds precedent and analogue in the fields of process control, material failure theory and fracture mechanics, where it is error, exception and failure that are the analytic objects of interest.
In material failure theory and fracture mechanics for example, scientists look for the fingerprint or signature of potential failure in the physical properties and brittleness of metal components. By analogy, we can think of settlement failure as being caused by the ‘economic brittleness’ of a dispute. But here, brittleness refers to the net predisposition to a fatal and permanent impasse which results from the economic structure of dispute and the trial prospects that confront each litigant.
The LSF Index provides a quantitative summary measure of this economic brittleness. It tells us when the potential for settlement has reached a breaking point or how close this might be.
Trial Selection Theory
Practitioners have long understood this connection between the economic factors of dispute and settlement decision making. In some respects the connection is an obvious one.
For example, we understand intuitively that as the anticipated costs of dispute become increasingly incurred or ‘sunk’ over time, the incentive to settle is reduced and with it the probability of a settlement.
Litigants will tend not to settle if there is little or nothing to be gained or saved by doing so. All things being equal, erosion of the cost surplus as a dispute progresses will act as an increasing impediment to settlement. We can easily demonstrate this using a game theoretic screening model of settlement bargaining. The chart below illustrates this cost-sinking effect for a typical dispute as the trial date approaches.
But costs are just one of many economic factors that influence settlement decision making and which as a consequence may have an impact on the probability of settlement and the risk of settlement failure.
In making settlement decisions, litigants compare their settlement option with their trial expectations and seek an optimum economic outcome as much as they do a just one. It is not difficult to see how many of the economic factors that affect this comparison (as they are anticipated or expected) will be influential to the probability of settlement. In this way, we can see how analysis of settlement failure risk is an analysis of the expectations about continued dispute.
The Forces of Settlement Failure
Many of economic factors, which comprise the expectations associated with continued dispute, will therefore exert a push or pull on the probability of settlement. And those which are influential will therefore contribute to the trial-or-settlement resolution of dispute in a categorical sense.
Where the combination of these economic forces favor agreement, a trial will be less probable and at some point a settlement may be reasonably predicted. Conversely, where the combination of forces are hostile to agreement, a trial will be more probable and at some point settlement failure may be reasonably predicted.
In this sense, we can see that disputes are ‘selected’ for trial based on their economic characteristics, the combination of which provide us with the identifying fingerprint of settlement failure risk.
Economic factors weigh on the likelihood of settlement failure – LSF Index is their combined effect
But several questions remain: Which are the factors that need to be considered? Will their impact on settlement failure risk be positive or negative? And how exactly do they combine with each other to create a net effect?
Fortunately, these questions have received considerable academic attention over the past 34 years. The branch of legal-economics that seeks to explain which cases are tried and which are settled is called trial selection theory.  
Unified Trial Selection Model
Canonical Trial Selection Models
In 1984, two separate and competing theories of trial selection began to evolve; each offering a different explanation as to why some disputes settle and some do not, and each outlining implications of the selection process itself for the trial rate and the implied plaintiff win rate.
These two canonical theories are known as divergent expectations (DE) and asymmetric information (AI).
DE theory suggests that litigants fail to agree a settlement because of excessive relative mutual optimism between the parties about their trial expectations. Whereas AI theory holds that cases fail to settle because of a high level of uncertainty in settlement bargaining about opponent trial expectations (this due to the presence of private information).
Although these two theories have been historically viewed as competing explanations, in reality both will have something to say about the risk of settlement failure in any individual legal dispute. Pairs of litigants can have varying degrees of relative mutual optimism or pessimism (‘optimism’). But they can also have uncertainty about their legal opponent (‘uncertainty’) – if only about their opponent’s reservation price.
To model the risk of settlement failure successfully, we first need to understand how the forces of optimism and uncertainty combine and interact with each other. That is, the two main theories of trial selection must be unified. And to be of practical relevance, a unified theory must allow us to describe the explicit relationship between each economic dimension of dispute and their contribution to the probability of settlement failure.
At SettlementAnalytics we have developed a proprietary approach to combining AI and DE theories into a single economic model. We call this the Unified Trial Selection Model™ (UTS Model™). In this combined model, note how optimism acts to amplify the settlement risk attributed to uncertainty (see Figure 2).
UTS Model allows us to compute the LSF Index for any given legal dispute at any point in time, and so achieve a quantitative insight into the likelihood of its settlement failure. The consequences of this innovation for litigants and law firms are exciting.
To begin with, we can now compute the LSF Index in the two dimensions of the AI and DE theories of selection. For the first time we can see how optimism and uncertainty combine to influence the probability of settlement and settlement failure for any particular lawsuit. What is revealed is a fascinating settlement failure probability surface, as illustrated above for a hypothetical dispute.
In Figure 2, we can see how high degrees of certainty (low uncertainty) and pessimism (negative optimism), combine to make settlement failure highly unlikely. Mutual pessimism creates a wider settlement bargaining zone within which litigants can find a negotiated solution, and high certainty reduces the incentive to take bargaining risk. Conversely, high optimism has the effect of collapsing the settlement bargaining zone making agreement less probable and high uncertainty encourages litigants to adopt more aggressive bargaining positions resulting in a greater probability that there will be a failure to agree.
Real Word Complexity
For any dispute, its optimism and uncertainty coordinates provide a rough sketch of its trial selection economic fingerprint. To estimate the chances of failing to reach a settlement, litigants should, at a minimum, consider the combined impact of AI and DE trial selection processes.
However, although this two-dimensional model offers a good start, there is much more to the economics of settlement failure than AI and DE theories allow. As we have already discussed, many of the economic factors that can influence a settlement decision may also have an impact on the probability of settlement and, therefore, a trial selection effect.
The chances that a dispute will settle can turn just as critically on things like costs, capital costs, time to trial, risk aversion, contingent fee structures, and cost rules – to name just a few. Research by SettlementAnalytics has identified over a dozen economic factors of dispute, all of which have a measurable influence on the prospects for settlement.
Whether we are modeling the trial rate across all disputes or the settlement probability for individual disputes, these many factors should be taken into account. It is their combined influence that will define the economic brittleness of a dispute and shape its settlement bargaining outcome.
For example, in one dispute, the forces of high award uncertainty, mutual optimism and low anticipated costs may be acting to increase the risk of a trial. At the same time, high risk aversion, a relatively high plaintiff cost of capital, and a split contingent fee structure may have an offsetting effect. In general, any combination of factors could exist, giving every legal claim a unique economic state at any point in time. But whatever the sign and scale of each settlement risk factor, it is their aggregate net effect that we need to capture. We can think of this aggregate net effect as expressing the net legal-economic vector which augurs for the failure to settle.
A major contribution of UTS Model and LSF Index is they incorporate all of the key financial and expectational dimensions of real-world legal claims in a unified AI-DE analytic framework. To our knowledge, LSF Index and UTS Model represent the most advanced, factor-rich analysis of settlement failure risk ever undertaken.
With LSF Index technologies, litigants and law firms can now quantitatively evaluate the likelihood of settlement failure for any dispute and monitor settlement risk over time as trial approaches. The schematic above describes our 3-step analytic process.
LSF Risk Levels & Trial Prediction
LSF Risk Levels
We divide the LSF Index range into five distinct LSF Risk Levels to provide a convenient color-coded barometer of settlement failure risk. Risk Levels capture the settlement failure risk reflected in the current Index level as well the typical acceleration of this risk as trial approaches.
Using the LSF Index as an input, we can now make a categorical trial-or-settlement outcome prediction for any individual legal dispute. Predictions involve the use of a threshold or cutoff value of the Index, above or below which cases can be categorized accordingly.
The cutoff value is chosen so as to correctly classify disputes with the highest level of accuracy. Cutoff values will be periodically updated using logistic regression techniques and having regard for the statistical measures of sensitivity and specificity, and the historical performance of the classification system.
Settlement Risk Engineering
With LSF risk analysis and prediction services, litigants and law firms can now identify trial-bound legal disputes before they become a problem. And, informed of a probable impending trial, timely trial-avoidance strategies can be considered.
Drawing on the application of LSF Index technologies, we introduce the practice of Settlement Risk Engineering, which we define as the purposeful and calibrated adjustment of the economic and informational landscape of dispute in order to push bargaining towards a settlement or a specific settlement target.
Informed of which factors are driving impasse and causing elevated settlement failure risk levels, litigants are better able to adjust (and/or signal) the factor inputs that are consistent with their legal bargaining aims.
A New Legal-Analytic Discipline
The Science of Settlement Failure
With the development of UTS Model and the launch of LSF Index, SettlementAnalytics has introduced a new legal-analytic discipline to the practice of litigation and claims portfolio management: Settlement Failure Risk Analysis and Prediction (‘SFRAP’).
After litigants have assessed the legal merits of a particular case and calculated its trial expected value in the usual way, they can now ask and answer important follow up questions, such as:
- Will this case settle?
- What is the probability of settlement failure?
- How will the expense of discovery impact settlement risk?
- What factor changes would be necessary in order to preserve a reasonable prospect of settlement?
- Which legal claims in a portfolio have the highest risk of proceeding to a court adjudication?
While SFRAP can be particularly helpful prior to the development of settlement strategy, periodic review and forecast over the course of a dispute can reveal important trends and alert litigants to a potential deterioration in the prospects for settlement.
With the LSF Index at hand, the tell-tale signs of impending settlement failure are finally revealed.
LSF Index Specifications
The LSF Index represent the theoretical probability of failing to resolve a dispute through settlement at any time over the remaining period prior to its trial (or arbitration) and adjudication. The Index is based on the Unified Trial Selection Model (UTS Model) developed by SettlementAnalytics. Index readings for any dispute reflect inputs provided to the Model by litigants and/or counsel.
The Index is standardized on a scale of 0 to 100. An Index level of 0 indicates there is, theoretically, a 0% probability of settlement failure, and a level of 100 corresponds to a 100% probability of settlement failure.
The Index level is dependent on and sensitive to inputs given to the UTS Model and is considered valid as at the time for which the inputs are current. As new information arrives during dispute, changes to the Model inputs will result in changes to the Index.
 In this brochure we will use the language of court adjudication, but the Index and the theory on which it is based apply equally to arbitration.
 For an example of a material failure index see, Tsai, S. W., and E. M. Wu., A General Theory of Strength for Anisotropic Materials, Journal of Composite Materials 5.1, 58-80, (1971).
 Model-based approaches to categorical prediction are subject to potential error. Appropriate caution should be exercised in application of prediction models and services. Practitioners should consult the LSF Index Disclosures referred to above.
 Coined by Hylton and Lin in 2009. See, Keith N. Hylton and Haizhen Lin, Trial Selection Theory and Evidence: A Review, Forthcoming in, Encyclopedia of Law and Economics: Volume X: Procedural Law and Economics, edited by Chris Sanchirico, Edward Elgar Publishing): 1-26, (September 2009), Web.
 Also referred to as case selection theory or selection theory.
 To our knowledge, LSF Index represents the first commercial application of trial selection theory.
 See, George L. Priest, and Benjamin Klein, The Selection of Disputes for Litigation, The Journal of Legal Studies 13, no. 1 (Jan., 1984).
 See, Bebchuk, Lucian Arye, Litigation and Settlement under Imperfect Information, The RAND Journal of Economics 15.3 (1984): 404; and Hylton, Keith N, Asymmetric Information and the Selection of Disputes for Litigation” Journal of Legal Studies, vol. 22, 187-210, (1993).
 There has already been academic progress in this direction. See, Keith N. Hylton and Haizhen Lin, Trial Selection Theory: A Unified Model, Boston University School of Law Working Paper No. 10-12 (May 17, 2010), Revision of May 2011, Web.
 The AI component of UTS Model draws on our fully-specified game theoretic models of signaling & screening. See: Robert Parnell, Conflict Analytics: The Game Theory of Legal Dispute, CMS Disputes Digest, Issue 5, 24, Oct 2016.
 Keith N. Hylton and Haizhen Lin, Trial Selection Theory and Evidence: A Review, Forthcoming in, Encyclopedia of Law and Economics: Volume X: Procedural Law and Economics, edited by Chris Sanchirico, Edward Elgar Publishing): 1-26, 2, (September 2009),Web.
 Hylton and Lin, 2009 (n 11), 14.
 Our research shows that the rate of change of the probability of a trial (the ‘trial risk delta’) typically increases as trial approaches.
 Framing the Index as an analysis of settlement failure should not be construed as attributing an inherent value to settlement over trial. Computation of the Index and trial selection theory in general ascribes no normative value to settlement in and of itself. For a thorough critique of the axiomatic belief in a normative superiority of settlement see, Robert J. Rhee, A Price Theory of Legal Bargaining: An Inquiry into the Selection of Settlement and Litigation under Uncertainty, 56 Emory L.J. 619, 620 (2006).
 While SFRAP is grounded in the ideas of trial selection theory, there is an important point of distinction. Trial selection theory has historically been focused on the implications of selection for the plaintiff win rate across the population of disputes (see n 16), whereas SFRAP is more directly concerned with the selection effect itself and its application to the analysis of individual disputes.
 Gross, Samuel R, Getting to No: A Study of Settlement Negotiations and the Selection of Cases for Trial, K. D. Syverud, co-author. Mich. L. Rev. 90, 319-93, 325 (1991).
 See Quantitative Settlement Bargaining Analysis™ (“QSBA”) and related courses at, https://settlementanalytics.com/services/quantitative-bargaining-analysis/