Strategic advocacy and advisory work in complex employment disputes, with particular expertise in discrimination, whistleblowing, TUPE, and appellate litigation.
Approach
Cases are won in cross-examination, not on the odds. But knowing the odds helps. Ten years at the Employment Bar. Before that: a doctorate, publishing, advocacy in Namibia, speeches for a Japanese Ambassador. Employment law is about people from all walks of life. Coming to the Bar the long way round has been good preparation for that.
Selected Instructions
Recent Articles
Analytical Toolkit
20 tools organised by stage of the litigation lifecycle — from initial viability assessment through to settlement negotiation and award analysis.
How to use these tools together
The tools are designed to chain together. The key insight is that probability of success from Stage 1 combines with maximum award from Stage 2 to produce an expected value that drives all Stage 3 analysis. The workflow below shows a typical progression from initial instruction to settlement negotiation.
The first check on any new instruction. Calculate the ET1 deadline accounting for ACAS Early Conciliation stop-the-clock rules. If the claim is out of time, no further analysis is warranted without a considered extension argument.
Structure every claim and sub-issue into a formatted document. For unfair dismissal, run the Disciplinary Checker simultaneously — procedural gaps directly affect both liability and any ACAS uplift claim in Stage 2.
Start with the Legal Element Analyser — feed each legal element (protected disclosure, causal link, detriment, etc.) through the relevant Bayesian network to get a doctrinal assessment of whether the legal tests are met. Take that Legal Merits score into the Case Success Predictor, which assesses one claim at a time: it combines evidence quality, jurisdictional strength, and an empirical prior drawn from HMCTS tribunal statistics for that claim type. Most cases involve multiple claims; assess your leading claim first, then repeat for any others that materially affect the litigation risk. Update the probability as disclosure arrives using the Evidence Updater. The resulting figure — p — is the single most important input to every Stage 3 tool.
Build the Schedule of Loss first to identify every applicable head. Then use the specialist calculators for the figures that require precise computation — pension loss in particular involves actuarial principles that are frequently disputed. The total is Vmax, the maximum achievable award at trial.
A single expected value can mislead. The Monte Carlo Settlement Range Modeller runs thousands of simulated tribunals varying each head within your min/mode/max estimates, producing P10/P50/P90 outcome bands. The Tornado diagram then identifies which uncertain factors produce the largest swing — focus pre-trial effort on resolving the top two or three.
When a settlement offer arrives, don't ask "is this fair?" — ask "what would p need to be for rejection to be rational?" The Threshold Calculator answers that question directly. The Decision Tree structures the full decision with BATNA (best-case alternative) and WATNA (worst-case alternative), showing the rational settlement zone as an overlap between both parties' positions.
The Ackerman Planner structures an opening anchor and declining concession sequence toward your target figure — anchoring high creates room to concede while still achieving your objective. For clients who are risk-averse (outcome variance matters more than EV), switch to Decision Analysis (regret minimisation), which weights the worst plausible outcome more heavily than the mathematical expected value.
The key chain: p (probability from Step 3) × Vmax (quantum from Step 4) = E[V] (rational settlement floor). Every offer below E[V] − litigation costs is, on expected-value grounds, worth rejecting. Every offer above it warrants serious consideration. The Threshold Calculator and Decision Tree make this explicit rather than intuitive.
Case Preparation
Begin here. Verify limitation before anything else. Then structure the live issues, assess each legal element, and calibrate your prior probability of success.
Quantum & Loss
Build and document the financial claim. Start with the Schedule of Loss, then use specialist calculators for pension loss, statutory payments, and interest.
Risk & Negotiation
Model the full distribution of outcomes, reverse-engineer settlement thresholds, and structure negotiation strategy using decision analysis and game theory.
Research & Analysis
Explore case law relationships, identify precedent patterns, and navigate complex liability structures across employment law topics.
Instructing Counsel
Resources and tools designed for employment law teams
All tools, calculators, and research materials on this site are for educational and informational purposes only. They do not constitute legal advice. Always verify calculations and guidance against primary sources and consult with counsel for specific matters.
Get in Touch / Resources
For instructions or advisory work, contact me directly or via St Philips clerks. Access key documents here:
Legal Research Database
Searchable database of employment law authorities with full summaries, judicial analysis, and cross-referenced statutes.
Explore Database →Practice Calculators
From limitation checks and schedules of loss through Monte Carlo simulation, decision trees, and Ackerman negotiation planning. Organised by stage of the litigation lifecycle.
View Tools →Strategic Advocacy
Complex discrimination, whistleblowing, TUPE, and appellate work. Court of Appeal experience. Available for trial and advisory.
View Experience →Quick Access Tools
🧠 Experimental Tools
What Clients Say
A superb and effective trial counsel.
Extremely responsive, knowledgeable and delivers high quality advice in a client-friendly manner.
Excellent preparation and advocacy that secured a successful outcome in a complex multi-day hearing.
Knowledge Base
Explore key areas of employment law and leading authorities.
This knowledge base is for educational and informational purposes only. It does not constitute legal advice. Case summaries and legal principles are simplified for reference and may not reflect the most current legal developments. Always consult a qualified professional for specific legal guidance.
List of Issues Builder
Generate a structured Word document (.docx) containing the statutory legal tests for your chosen claims.
This is an experimental educational tool for illustrative purposes only. It does not constitute legal advice. The generated document contains abstract legal tests which must be populated with specific factual allegations. Always consult a qualified professional for specific legal guidance.
A Tribunal will likely reject a List of Issues that only recites the abstract legal tests. To be useful, you must edit the downloaded document to populate it with your specific factual allegations (e.g., dates of incidents, specific words used, names of comparators).
Select Claims & Jurisdictions
Equality Act 2010
Employment Rights Act 1996 — Dismissal
ERA 1996 — Detriment, Wages & Other
TULR(C)A 1992
TUPE & Regulations
Common Law & Contract
Remedy
Schedule of Loss Builder
Generate a structured Word document (.docx) containing the framework for a Claimant's draft Schedule of Loss. Select the relevant heads of loss and enter key case details.
This is a practice tool for illustrative purposes only. It does not constitute legal advice. The generated Schedule of Loss contains structural headings and placeholder calculations which must be populated with specific figures, evidence references, and supporting narrative. Always consult a qualified professional.
A Schedule of Loss is a living document that should be updated as the case progresses to reflect developments in mitigation, updated losses, and disclosure. The downloaded draft provides a structural framework only — you must edit it to include specific figures, mitigation details, evidence references, and any other case-specific narrative before it is fit for use.
Key Case Details
Select Heads of Loss
Unfair Dismissal
Notice Pay / Wrongful Dismissal
Discrimination / Whistleblowing
Wages & Other Losses
Recent Instructed Cases
News & Insights
Articles and insights on this site are for educational and informational purposes only. They do not constitute legal advice. Content may not reflect the most current legal developments. Always consult a qualified professional for specific legal guidance.
Video Insights
Short video commentaries on recent appellate decisions and key employment law developments.
Video content is for educational and informational purposes only. It does not constitute legal advice. Discussions of cases and legal principles are simplified for accessibility. Always consult a qualified professional for specific legal guidance.
Coming Soon
Video commentaries on significant EAT and Court of Appeal decisions will appear here. Subscribe to be notified when new content is published.
Holiday Pay
For irregular-hours workers (52-week reference period estimate).
This is a practice tool for illustrative purposes only. It does not constitute legal advice. Calculations are based on standard 52-week reference periods and may not account for complex patterns of work or specific contract terms. Always consult a qualified professional for specific legal guidance.
📐 Mathematical Basis
• $\bar{W}$ = Average weekly pay over reference period
• $P_i$ = Gross pay in week $i$
• $n$ = Number of weeks worked (max 52)
• 5.6 = Statutory annual leave entitlement in weeks
Limitation Date Calculator
Calculate limitation deadlines for Employment Tribunal claims and EAT appeals.
This is a practice tool for illustrative purposes only. It does not constitute legal advice. Limitation dates are complex and may be affected by specific facts not captured here. Missing a deadline has severe legal consequences. Always consult a qualified professional for specific legal guidance.
📐 Mathematical Basis
• The primary limitation period is 3 months less one day from the effective date of termination (EDT) or the date of the act.
• ACAS Early Conciliation "stops the clock" on this period, extending it as per statutory rules (see below).
- Employment Rights Act 1996, s.111(2) (Unfair Dismissal)
- Equality Act 2010, s.123 (Discrimination Claims)
- The Employment Act 2002 (Dispute Resolution) Regulations 2004 (for ACAS early conciliation rules)
📐 Legal Basis
Rule 3(3) of the EAT Rules sets a 42-day time limit for lodging an appeal. The day the document is sent/dated does not count (Practice Direction 2024, para 3.4.1). The deadline expires at 4pm on the final day (rule 37(1A)).
Judgments: Time runs from the date written reasons are sent to the parties (rule 3(3)(a)(i)). Where written reasons were not requested, time runs from when the written record of the judgment was sent (rule 3(3)(a)(iii)).
Orders: Time runs from the date of the order itself (rule 3(3)(b)), not the date it is sent.
Bank Holidays: If the deadline falls on a day the EAT office is closed (weekends, bank holidays), it is extended to the next working day (rule 37(2); Practice Direction 2024, para 3.4.3).
- EAT Rules 1993, rule 3(3) (as amended)
- EAT Practice Direction 2024, paras 3.3–3.4
- Gdynia American Shipping Lines v Chelminski [2004] ICR 1523, CA (time runs from date of sending)
- Rana v LB Ealing [2019] ICR 789, CA (mis-sending; discretion to extend)
- Kennaugh v Lloyd-Jones EAT 0710/07 (certificate of correction)
Statutory Notice Pay Calculator
This is a practice tool for illustrative purposes only. It does not constitute legal advice. It calculates statutory minimum notice based on years of service and does not account for enhanced contractual notice periods. Always consult a qualified professional for specific legal guidance.
📐 Mathematical Basis
• $Y$ = Completed years of service
• Gross Weekly Pay is subject to statutory cap.
- Employment Rights Act 1996, s.86 (Statutory minimum period of notice)
- Employment Rights Act 1996, s.220-229 (Calculation of a week's pay, including statutory cap)
Redundancy Pay Calculator
This is a practice tool for illustrative purposes only. It does not constitute legal advice. It calculates statutory redundancy pay based on age, service and pay (capped). It does not account for enhanced contractual redundancy. Always consult a qualified professional for specific legal guidance.
📐 Mathematical Basis
• $S_{total}$ = Total completed years of service (max 20 years used for calculation)
• Weekly Pay is subject to the statutory cap.
• Age is at the effective date of termination.
- Employment Rights Act 1996, s.162 (Amount of a redundancy payment)
- Employment Rights Act 1996, s.227 (Limit on amount of week's pay)
- Employment Rights Act 1996, s.228 (Maximum number of years' service)
Procedural Fairness Checker
This is a practice tool for illustrative purposes only. It does not constitute legal advice. It provides a basic checklist for ACAS-compliant disciplinary procedures. Procedural fairness depends on the full circumstances of each case. Always consult a qualified professional for specific legal guidance.
View the full ACAS Code →
ACAS Code Compliance Checklist
Investigation
Notification
The Hearing
Decision & Warnings
Appeal
ACAS Guidance
Settlement Negotiation Planner
Plan a structured sequence of offers designed to reach your target settlement without splitting the difference.
This tool generates a disciplined offer sequence based on the Ackerman/Voss protocol. You set the target figure; the tool calculates when and how much to concede at each round. It does not constitute legal advice.
The Ackerman model avoids the common trap of "splitting the difference". Instead, you make a series of carefully planned offers, each one smaller than the last, so your opponent sees you approaching a firm limit. Enter your target settlement figure, choose your perspective (paying or receiving), and optionally enter the opponent's opening position to see where the two sides are likely to converge.
🤝 How the Offer Sequence Works
Convergence Plot: Shows how your offers move towards your target settlement. If you've entered the opponent's position, their projected concessions appear as a dashed line. Where the two lines meet is the likely settlement zone.
Negotiation Surface: A 3D visualisation of negotiation "tension". The saddle point (the flat middle) represents the equilibrium where neither side can improve their position without the other walking away — this is your target settlement.
Interest on Employment Tribunal Awards
Calculate simple interest at 8% per annum on tribunal awards under the Employment Tribunals (Interest) Order 1990 and the Employment Tribunals (Interest on Awards in Discrimination Cases) Regulations 1996.
This is a practice tool for illustrative purposes only. It does not constitute legal advice. Interest calculations may be affected by specific facts not captured here, including the tribunal's discretion under Reg 6(3) to vary dates where serious injustice would result. Always consult a qualified professional for specific legal guidance.
📐 Mathematical Basis
• \(I\) = interest payable
• \(P\) = principal (award amount)
• \(r\) = annual interest rate (currently 0.08, i.e. 8%)
• \(d\) = number of days in the interest period
Interest period:
• Injury to feelings — from the date of the act of discrimination to the day of calculation (Reg 6(1)(a)).
• Financial loss — from the “mid-point date” to the day of calculation (Reg 6(1)(b)). The mid-point date is the date halfway through the period beginning on the date of the act of discrimination and ending on the day of calculation (Reg 4(2)).
- Employment Tribunals (Interest on Awards in Discrimination Cases) Regulations 1996 (SI 1996/2803), Regs 2–6
- Employment Tribunals (Interest) Order 1990 (SI 1990/479), Arts 3–4
- Judgments Act 1838, s.17 (fixing the rate at 8%)
- Equality Act 2010, s.139 (power to make subordinate legislation on interest in discrimination cases)
- Komeng v Creative Support Ltd EAT 0275/18 — tribunals must consider awarding interest regardless of whether the parties apply for it
- Marshall v Southampton and South West Hampshire Area Health Authority [1993] ICR 893, ECJ — interest is an essential component of discrimination compensation
Case Success Predictor
A structured method for quantifying your confidence in a particular cause of action — combining your assessment of jurisdictional strength, legal merits, and evidence quality with an empirical baseline drawn from HMCTS tribunal statistics.
The tool assesses one claim at a time. Most employment cases involve several claims running in parallel — unfair dismissal, discrimination of one or more types, breach of contract, and so on — each with its own legal elements, strength, and statistical baseline. This tool cannot model that interaction. The practical approach is to run the assessment on your leading claim: the one that drives the litigation risk, the remedy sought, and the settlement calculus. You can repeat the exercise for any other claim that matters. The three factor scores reflect your professional judgment; the historical baseline is derived from verified HMCTS hearing success data by claim type. You control the weight given to each factor; the tool combines them and shows its working.
This tool structures professional judgment on a single cause of action — it does not replace that judgment, and it does not model the cumulative effect of multiple claims. The factor scores you set must reflect your informed assessment of the specific claim being evaluated. It is not legal advice.
Data source & methodology: Rates represent the hearing success rate — the proportion of claims determined at a full merits hearing where the claimant succeeded — derived from HMCTS Employment Tribunal Statistics 2019/20 (four-quarter average from official HMCTS ET_3 tables). They exclude cases settled by ACAS, withdrawn, struck out, or disposed of by default judgment, so they are not overall success probabilities. HMCTS stopped publishing jurisdiction-specific disposal data in September 2022 following migration to a new case management system; these are the most recent verified figures available from official sources.
Known data quality issues: Three claim types require caution. Redundancy Pay (79.0%) and Working Time (74.8%) have high hearing success rates reflecting that claims reaching hearing tend to be factually strong; the prior should be treated sceptically absent that selection effect. Equal Pay (54.5%) and Sexual Orientation (25.5%) are derived from very small quarterly case volumes and are statistically unreliable as priors. Reduce the Historical Rate Weight slider toward zero for these claim types and rely primarily on your factor assessments.
How confident are you that limitation and jurisdictional arguments will be resolved in the claimant's favour? The outcome is binary — a claim either proceeds or it does not — but the practitioner's pre-decision assessment is not. Multiple sources of uncertainty commonly bear on this:
- Just and equitable extension (s.123(1)(b) EqA) — a broad, fact-sensitive discretion; far from automatic
- Not reasonably practicable extension (s.111(2) ERA 1996) — narrower and harder to establish
- Continuing acts — whether a series of acts constitutes conduct extending over a period or a string of discrete (some time-barred) acts is frequently contested
- Date disputes — effective date of termination, when constructive dismissal was accepted, or when the last act occurred may all be argued
- ACAS early conciliation — miscalculation of the extended deadline, or ambiguity in the conciliation timeline, can place a claim in genuine uncertainty
- Wrong respondent / late addition — adding a respondent after primary limitation has expired engages Rule 34 ET Rules; not guaranteed
- Employment status — where jurisdiction depends on worker or employee status, a disputed status argument undermines the jurisdictional foundation entirely
How strong is the legal argument? Consider whether each element of the cause of action can be established, the quality of any comparator, and whether the respondent has viable defences (e.g., justification, fair reason).
How strong is the documentary and witness evidence? Consider contemporaneous records, credibility of witnesses, disclosure risks, and whether the burden of proof shifts (s.136 EqA).
How much should the general tribunal success rate for this claim type anchor the prediction? A low weight means your case-specific factors dominate; a high weight pulls toward the statistical average.
How It Works
The model starts from the historical tribunal success rate for the selected claim type, then adjusts up or down based on three case-specific factors you assess: jurisdiction, legal merits, and evidence quality.
Weights control how much each factor influences the overall prediction. They must total 100%. Adjusting one weight automatically redistributes the others.
The result is only as good as the assessments you put in. Two experienced practitioners may reasonably differ on factor scores — the tool makes that disagreement visible and structured rather than hidden in a gut feel.
📚 Mathematical Foundation: Weighted Linear Scoring Model
This tool uses a weighted linear scoring model — additive in structure, sharing the form of the General Linear Model (GLM), but with an important distinction: the factor weights ($w_i$) are set by the practitioner, not estimated from a regression dataset. The historical baseline $H$ is an empirically-derived prior drawn from HMCTS ET hearing success data disaggregated by claim type; the remaining terms adjust it based on case-specific professional assessment.
This model: $H$ is the empirically-grounded intercept — the historical hearing success rate for the selected claim type, derived from HMCTS ET_3 tables. The three factor terms adjust upward or downward from that baseline. Each factor score is centred at 50, so a neutral assessment (50) contributes zero adjustment; scores above 50 improve the estimate, scores below 50 reduce it.
• $H$ = Historical hearing success rate by claim type (HMCTS ET_3 data)
• $w_i$ = Normalised weight for factor $i$, set by the practitioner
• $S_i$ = Factor score (0–100), centred at 50 (neutral)
• Subscripts: L = Limitation/Jurisdiction, B = Legal Merits, E = Evidence
💰 Next Step: Settlement Analysis
Model the financial settlement range and litigation risk using this success probability.
🌪️ Which Factor Matters Most?
Test which variable drives the prediction — focus preparation where it has the most impact.
⚖️ Legal Element Analyser
Examine the logical structure of a legal claim — element by element. Does the disability exist? Does something arise from it? Was the treatment unfavourable? Each limb depends on the last, and this tool makes those dependencies explicit.
This tool models how legal elements depend on each other. Each input represents your professional assessment of whether a legal test is met. The network calculates the combined probability using Bayesian inference — propagating your assessments through the causal chain to the outcome node.
Path Dependency: These networks enforce sine qua non logic. Where a legal element is a prerequisite for downstream elements (e.g., disability must be established before "something arising" can exist under s.15 EqA 2010), the conditional probability is locked at 0%. Sliders marked as locked reflect legal impossibility, not mere improbability.
Employment claims require proving a chain of legal elements — each building on the last. For example, a s.15 disability discrimination claim needs: (1) a disability, (2) something arising from it, (3) unfavourable treatment, and (4) no justification. This tool lets you set your confidence on each element and instantly see the combined probability. It is purely doctrinal: it asks whether the legal test is met on your assessment of the facts, and says nothing about evidence quality, tribunal statistics, or jurisdictional risks. For a composite assessment that layers in those factors, take your Legal Merits result to the Case Success Predictor.
📚 Mathematical Foundation: Bayesian Networks
This tool utilizes Bayesian Networks (BNs), a powerful probabilistic graphical model that represents a set of random variables and their conditional dependencies via a directed acyclic graph (DAG).
• $X_i$ = A random variable (node) in the network
• Parents($X_i$) = The set of parent nodes directly influencing $X_i$
• $P(X_i | \text{Parents}(X_i))$ = Conditional Probability Distribution (CPD) for $X_i$ given its parents.
Inference in Bayesian Networks, the process of computing the posterior probability of a variable given evidence, is performed using algorithms like Variable Elimination or Message Passing.
• Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann.
• Koller, D., & Friedman, N. (2009). Probabilistic Graphical Models: Principles and Techniques. MIT Press.
🔄 Evidence Updater
See how new evidence shifts your assessment of case strength or quantum.
Start with your initial assessment, then input the new evidence. The tool calculates a mathematically weighted update, giving more weight to whichever input you're more confident about.
Litigation assessments change as evidence emerges. A case you rated at 60% before disclosure might shift to 40% after seeing the documents — or to 80%. This tool formalises that process: enter your starting view and the new evidence, and see exactly how the updated assessment is calculated. Use Success Likelihood mode for probability of winning; use Quantum Estimate mode for damages values.
📚 Mathematical Foundation: Beta-Binomial Conjugate Updating
This mode implements Bayesian inference for probabilities, using the Beta-Binomial conjugate model. Your starting assessment is encoded as a Beta distribution, and evidence is incorporated via exact conjugate updating (Bayes, 1763; Laplace, 1774).
• $P(H)$ = Prior probability — your initial belief before seeing evidence
• $P(E \mid H)$ = Likelihood — probability of observing evidence $E$ if hypothesis $H$ is true
• $P(E)$ = Marginal likelihood — total probability of evidence under all hypotheses
• $P(H \mid E)$ = Posterior probability — updated belief after observing evidence
Why Beta-Binomial? Probability of success is naturally bounded between 0 and 1. The Beta distribution lives on exactly this interval, making it the correct conjugate prior for binary-outcome reasoning (win/lose). The posterior is computed in closed form and credible intervals are derived from Beta quantiles (the regularised incomplete beta function).
📚 Mathematical Foundation: Normal-Normal Conjugate Updating
This mode uses Normal-Normal conjugate updating — the Bayesian method for combining two uncertain estimates of a continuous quantity such as damages. Rather than simply averaging two point estimates, it weights each by its precision (the inverse of variance), so the more confident estimate pulls harder.
• $\mu_0$ = Prior mean — your initial quantum estimate (£)
• $\sigma_0$ = Prior uncertainty — how widely your estimate could vary (± £)
• $\bar{x}$ = Evidence value — what the new evidence suggests the quantum is (£)
• $\sigma_E$ = Evidence uncertainty — how precise that evidence-based valuation is (± £)
• $\mu_{\text{post}}$, $\sigma_{\text{post}}$ = Updated estimate and uncertainty after combining both sources
Why Normal-Normal? Unlike probabilities (which are bounded 0–1 and require a Beta distribution), monetary quantum is a continuous, unbounded quantity. The Normal distribution is the natural model for such estimates, and when both prior and likelihood are Normal, the posterior is also Normal in closed form — a conjugate update. This is exact (no simulation or approximation required) and the credible interval is computed directly from the posterior standard deviation.
• Bayes, T. (1763). An Essay towards solving a Problem in the Doctrine of Chances. Phil. Trans. Royal Soc. 53:370-418.
• Laplace, P.S. (1774). Mémoire sur la probabilité des causes par les événemens. Mém. Acad. Roy. Sci. 6:621-656.
• Gelman, A. et al. (2013). Bayesian Data Analysis (3rd ed.). Chapman & Hall/CRC.
⚖️ Pension Loss Calculator
Calculate the pension element of compensation following the Principles for Compensating Pension Loss (4th Ed, 2021).
Pension loss calculations require actuarial understanding. This tool implements the judicial guidance methodology. Always verify against the full Ogden Tables and current discount rates.
Load a pre-filled example from Appendix 3 of the judicial guide, or enter your own values below.
A pot-based pension — your employer pays a set % of salary into an investment pot (e.g. NEST, NOW: Pensions, most auto-enrolment schemes). You're calculating lost employer contributions.
A 'final salary' or 'career average' pension — promises a guaranteed annual income in retirement (e.g. NHS Pension, Teachers' Pension, Civil Service). Calculation requires Ogden Table multipliers.
For DC pensions, the loss is straightforward: the employer contributions that would have been paid into the pension pot during the loss period.
Annual Pension Values
Actuarial Factors (Ogden Tables)
Lump Sum Adjustments
⚖️ Settlement Range Modeller
Model the range of possible tribunal awards by running thousands of simulated outcomes.
Enter your best, worst, and most likely estimates for each head of loss. The tool runs thousands of simulated tribunals to show the probable range of total awards.
The output is a distribution of composite payoffs, not a single number. For each head of loss you select, provide three estimates (minimum, most likely, maximum). The tool independently samples each head thousands of times and sums them to build a realistic picture of total outcomes. You can also factor in estimated legal costs as a deduction, so the result shows your net position after representation costs. This helps frame realistic settlement discussions by showing the range of probable awards rather than a false-precision single figure.
📚 Mathematical Foundation: Monte Carlo & PERT
This tool models litigation uncertainty using Monte Carlo simulation. It performs thousands of trials, sampling from probability distributions for each head of loss to generate a range of potential outcomes.
By aggregating these distributions (e.g., Unfair Dismissal + Vento + Pension), the model reveals the probability density of the total award, helping to identify the most likely settlement range.
• Malcolm, D. G., et al. (1959). Application of a Technique for R&D Program Evaluation (PERT). Operations Research.
• Vose, D. (2008). Risk Analysis: A Quantitative Guide. John Wiley & Sons.
• Palisade Corporation. (2024). Guide to Monte Carlo Simulation.
Sensitivity Analysis
Identify which factor drives your case assessment most. Toggle individual variables to see which has the greatest impact on the overall prediction — then focus preparation on the load-bearing elements.
The tornado diagram visualises which factor, if it swings from its worst to best plausible value, moves the needle most. Tall bars = high leverage. Short bars = marginal impact.
This tool structures professional judgment — it does not replace it. Set your baseline assessments, then define plausible low and high values for each factor. The diagram reveals which uncertainties matter most. It is not legal advice.
For each factor, set your baseline assessment (your best estimate), then define the low (pessimistic) and high (optimistic) plausible values. The tornado diagram shows the swing each factor produces.
Weights must total 100%. Adjust how much each factor influences the overall prediction.
Set your baseline and range values, then click Generate Tornado Diagram to see which factor has the most impact.
How It Works
The tornado diagram tests each factor one at a time. It holds all other factors at their baseline values, then swings one factor from its low to its high estimate. The resulting change in the overall prediction is plotted as a horizontal bar.
Red bars show the downside swing (baseline to low). Green bars show the upside swing (baseline to high). Longer bars mean that factor has more leverage over the final result.
Use this to prioritise case preparation: if evidence quality produces the widest swing, that's where to focus resources. If limitation is narrow, it's already secure.
📚 Mathematical Foundation: One-Way Sensitivity Analysis
Sensitivity analysis is a core technique in decision analysis and operations research. The tornado diagram implements one-way sensitivity analysis: each input variable is varied independently while all others are held constant (ceteris paribus).
• $\Delta_i$ = Total swing range for factor $i$
• $P(x_i^{\text{high}})$ = Prediction when factor $i$ is at optimistic value
• $P(x_i^{\text{low}})$ = Prediction when factor $i$ is at pessimistic value
• $H$ = Historical baseline rate for claim type
• Clemen, R. T. & Reilly, T. (2014). Making Hard Decisions with DecisionTools. Cengage Learning. Ch. 5 (Sensitivity Analysis).
• Eschenbach, T. G. (1992). Spiderplots versus Tornado Diagrams for Sensitivity Analysis. Interfaces, 22(6), 40–46.
• Howard, R. A. & Abbas, A. E. (2016). Foundations of Decision Analysis. Pearson. Ch. 14.
🎯 Next Step: Threshold Analysis
Reverse-engineer the problem: how confident must you be in each factor for a settlement offer to be worth rejecting?
Threshold Calculator
Reverse-engineer the decision: how confident must you be in a given factor for a settlement offer to be worth rejecting?
Instead of asking "What is our chance of winning?", this tool asks the harder question: "Given this settlement offer, what minimum confidence in each factor makes trial the rational choice?" If the required confidence exceeds what the evidence supports, the offer may be the better decision.
This tool helps structure settlement decisions — it does not make them. Legal costs, delay, emotional toll, and enforcement risk are not captured here. Always apply professional judgment. It is not legal advice.
Your current best estimates for each factor (these are your baseline — the tool will calculate what they'd need to be).
The amount offered to settle. The tool calculates the minimum success probability needed to justify rejecting this offer.
Your best estimate of the total tribunal award (including all heads of loss) if the claim succeeds. Use the Settlement Range Modeller median if available.
Total estimated costs of proceeding to hearing. Enter 0 if publicly funded or no representation costs.
Factor-by-Factor Threshold
For each factor: the minimum score needed (holding others constant) for trial to have a higher expected value than the settlement offer. Bars show your current assessment against the required threshold.
How It Works
The break-even probability is the point where the expected value of trial equals the settlement offer. Below this probability, accepting the offer is rational. Above it, trial has a higher expected value.
The factor thresholds reverse-engineer this: for each factor, what score would it need to reach (holding the others at your current assessments) to push the overall probability above break-even?
If a required threshold is above 100% or below 0%, it means that factor alone cannot push the result past break-even — it depends on the other factors too.
📚 Mathematical Foundation: Expected Value & Threshold Inversion
The tool finds the indifference probability $p^*$ where the expected value of litigation equals the certain settlement value, then inverts the prediction model to find factor-level thresholds.
• $p^*$ = Break-even probability (indifference point)
• $S$ = Settlement offer amount
• $A$ = Expected award if successful
• $C$ = Estimated legal costs to trial
• $S_i^*$ = Minimum required score for factor $i$
• $H$ = Historical baseline rate for claim type
• Raiffa, H. (1968). Decision Analysis: Introductory Lectures on Choices Under Uncertainty. Addison-Wesley.
• Mnookin, R. H. (2000). Beyond Winning: Negotiating to Create Value in Deals and Disputes. Harvard University Press.
• Guthrie, C. (2003). Prospect Theory, Risk Preference, and the Law. Northwestern University Law Review, 97(3), 1115–1163.
🌪️ See What Drives the Result
Use the Sensitivity Analysis tool to visualise which factor has the most leverage over your case prediction.
Case Law Network
Interactive map of 500+ employment law authorities, clustered by topic.
This is an experimental educational tool for illustrative purposes only. It does not constitute legal advice. Clusters and connections represent topic shared between cases and are not a substitute for legal research. Always consult a qualified professional for specific legal guidance.