Reading a fact pattern
Issue spotting looks like intuition from the outside. It isn't. It is a sequence of decisions, each of which can be named, taught, and practised separately — and most of what separates a confident reader from a stuck one is knowing that the sequence exists.
This page names the decisions. It assumes no prior familiarity with how law school works, and it does not assume you have anyone at home to ask.
- Start with the task, not the story
- The decomposition routine
- Facts, inferences, and conclusions are different objects
- Turning a topic into an issue
- The grammar of a rule
- The inferential bridge
- Using cases by relationship, not resemblance
- One routine, seven subject structures
- When you are stuck, name the failure
1. Start with the task, not the story
Read the call of the question first, then read the whole pattern once without stopping to analyse. You are looking for four things before you look at anything else.
- The requested product. Liability, ownership, a remedy, the outcome of a motion, advice to a client, or an evaluation of the policy. These are different deliverables and a strong answer to the wrong one earns nothing.
- Whose position must be analysed. "Discuss the café's exposure" and "discuss all parties' rights" are different assignments.
- The governing law and any exclusions. A torts question may deliberately exclude contract remedies. Following the scope is itself a graded skill.
- The time and the procedural stage. What a court can do at a pleading stage is not what it can do at trial.
The instinct to show everything you know is the single most expensive habit in a timed exam. Coverage of what was asked beats coverage of what you studied.
2. The decomposition routine
Ten questions, in a loop rather than a line. You will often reach step ten and discover a missing fact that sends you back to step two. That is the routine working, not failing.
| Step | The question | What you produce |
|---|---|---|
| 1. Actors | Who acted, and in what role? | People and entities, with roles: owner, employee, buyer, tenant |
| 2. Chronology | What happened before what? | A timeline separating promises, conduct, notice, and harm |
| 3. Interests | What is being protected or sought? | Bodily integrity, reliance, payment, possession, use, process |
| 4. Relationships | Who may assert what against whom? | Separate claimant–respondent–theory links |
| 5. Framework | Which body of law supplies the test? | A source and jurisdiction, or the rule the question supplies |
| 6. Rule structure | What must be established, and how do the parts connect? | Elements, alternatives, factors, exceptions, burdens |
| 7. Fact mapping | Which facts support, weaken, or leave each requirement uncertain? | Fact-to-element connections, with directions |
| 8. Reasoning | Why does that connection follow? | An inference, analogy, distinction, or interpretation |
| 9. Opposition | What is the strongest competing account? | A counterargument, defence, exception, or alternative rule |
| 10. Consequence | What follows, and with what confidence? | A qualified conclusion and any outcome-changing missing fact |
Steps one through four are reading. Steps five and six are knowledge. Steps seven through ten are the analysis everyone means when they say "apply the law to the facts." Most people who feel stuck at step seven are actually missing step six, and no amount of practice at step seven will repair that.
3. Facts, inferences, and conclusions are different objects
A pattern hands you a mixture: things stipulated as true, things a reasonable person might draw from them, things a party is asserting, and legal characterisations. Treating all four as the same object is the most common early error, and it produces answers that assume what they were supposed to argue.
Take a two-sentence pattern: The manager received an email at 10:00. The shopper fell at 10:30.
| Statement | What it is | Why the difference matters |
|---|---|---|
| The email was received at 10:00 | Express fact | Receipt is given. Reading it is not. |
| The manager had an opportunity to respond | Inference | Depends on content, availability, and circumstances |
| The manager knew about the hazard | Possible inference | Do not get from receipt to knowledge for free |
| The manager breached a duty | Legal conclusion | Requires a duty and an assessment of the conduct |
| The manager is liable | Ultimate conclusion | Other elements and defences still stand in the way |
Two habits follow. An unstated fact is unknown, not false — mark the gap rather than filling it. And a party's denial is a position, not evidence: the café saying the fall would have happened anyway creates a dispute, it does not win one. The reverse also holds. If the pattern stipulates a fact, do not manufacture a dispute about it to avoid applying the rule.
Which kind of object is this?
Same pattern each time: Quinn tells the manager at 9:00 that the handrail is loose. The manager keeps the entrance open. At 10:00 the rail detaches and a customer falls.
4. Turning a topic into an issue
"Negligence" is a topic. "The store was negligent" is a conclusion. An issue is the question in between: a disputed point that the governing rule requires someone to resolve, phrased so that it has an answer.
The test for whether you have an issue rather than a topic: can it be answered yes, no, or genuinely uncertain? "Negligence" cannot. "Whether an hour between notice and injury permitted a warning" can. Notice also that the sub-issue is where the facts actually live — the interval, the available precautions — which is why analysis that stops at the issue level always reads thin.
5. The grammar of a rule
Rules are not all checklists, and reading every rule as a checklist is why some analysis comes out mechanical and wrong. The connectors carry as much law as the words they join.
| Structure | How to analyse it | The frequent mistake |
|---|---|---|
| A and B and C | Establish every element | Letting strong proof of A cover a missing C |
| A or B | Analyse the applicable branch; sometimes both are live | Demanding both |
| A unless E | Establish the rule, then the exception and who must prove it | Forgetting the exception, or guessing at its burden |
| Multiple factors | Explain direction and weight under the governing test | Counting factors as if each were separately required |
| Threshold, then merits | Resolve the gateway before the downstream question | Using a strong merits position to skip the gateway |
| A presumption | Identify the triggering facts and its legal effect | Treating it as unrebuttable |
| A standard ("reasonable") | Apply contextual facts and precedent | Restating "reasonable" as though it explained something |
Watch the small words too. "Notice before acceptance" contains a temporal relationship; finding both events does not resolve it. "Substantial and unreasonable" is two requirements wearing one phrase.
What shape is this rule?
Read for the connectors. The subject matter does not matter here; the architecture does.
6. The inferential bridge
This is the part that gets graded and the part most often skipped. A fact and an element sitting next to each other in a sentence is not analysis. The analysis is the explanation of why the fact bears on the element, and what it does not establish.
"There was notice, so there was negligence."
"The warning email may support prior notice. If it identified this hazard and reached the responsible manager in time, a factfinder could treat the failure to warn customers as unreasonable. Receipt alone does not establish that the manager read it, and the opportunity to respond remains contested."
The second version is not better because it is longer. It is better because it names what the fact supports, what it does not reach, and what the other side gets to say. Note also that the conclusion stays proportionate: "could treat as unreasonable," not "was unreasonable." Overclaiming is its own error, and graders see it as a failure to understand the fact rather than as confidence.
Which version earns the point?
All four say something true. Only one builds the bridge.
7. Using cases by relationship, not resemblance
The temptation with an assigned case is to look for a new pattern that feels like it. Surface similarity is the weakest kind of analogy and graders can see it instantly.
Suppose an assigned case finds inadequate warning where an employee knew of a hazard for a substantial interval. A new pattern also happens in a supermarket, but supplies no notice at all. The shared setting is decoration. The missing notice is decisive, and a warehouse case with the same notice structure is the closer analogy despite looking nothing alike.
So the question to ask is never "what does this remind me of." It is: which shared feature matters under the rule, and does the new pattern have it?
8. One routine, seven subject structures
The decomposition routine does not change between courses. What changes is the organizing question you ask at step five, and the shape of the record you build. These are navigation prompts — you populate them with your own assigned authorities and your jurisdiction's tests. No row is a complete checklist.
| Subject | Organizing question | Decomposition | Typical trap |
|---|---|---|---|
| Torts | Who invaded whose protected interest, through what conduct? | Each claimant and defendant; theory; duty or intent; conduct; causation; harm; defences; remedies | Harm as a shortcut to liability |
| Contracts | What obligations became enforceable, and what happened to performance? | Governing law; formation; enforceability; terms; conditions; breach; excuses; remedies | Reaching breach before identifying the obligation |
| Property | Who holds which right, in what thing, at what time, against whom? | Asset; source of interest; transfers; possession and use; duration; notice and priority; remedies | Treating possession and title as the same thing |
| Civil procedure | Can this court resolve this dispute through this mechanism? | Court authority; parties; service; stage; motion; standard; record; requested order | Applying trial proof standards at the pleading stage |
| Criminal law | Has the prosecution established the charged offence under its governing law? | Conduct; circumstances; result; mental state per element; causation; defences; burden | Inferring culpability from a bad result |
| Constitutional law | What government action is challenged, through which provision? | Justiciability; actor; power or right; applicable test; facts; justification; relief | Choosing a level of scrutiny before framing the question |
| Evidence | What item is offered, by whom, and for what purpose? | Proposition; relevance; exclusion rules; hearsay purpose and exceptions; foundation; limits | Calling a statement hearsay without asking why it is offered |
A worked contrast, in three sentences
Torts. Build a party-by-party map before choosing a theory. Not every injury is negligence and not every deliberate act is a battery; the conduct picks the theory, not the outcome.
Contracts. Reconstruct the transaction before judging the failure. Proposals, responses, conduct, changed terms, performance, notice, refusal — in order, because a formation question answered out of sequence answers the wrong thing.
Property. Track rights through time in a table: asset, holder, right, source, start, end, competing claimant. Someone can own a future interest now without any present right to possess, and a tenant's possession says nothing about who holds the fee.
9. When you are stuck, name the failure
"I'm bad at issue spotting" is not a diagnosis and cannot be repaired. These can. Find the row that matches what actually happens when you sit down with a pattern, and do the thing in the third column instead of doing more full problems.
| What happens | What it usually means | What to do instead |
|---|---|---|
| You cannot explain a word in the rule | Vocabulary, not reasoning | Definition, example, and a near-miss non-example |
| You know the words but drop a requirement | Rule structure | Rebuild the rule from its elements and connectors |
| You list every doctrine you know | Relevance and discrimination | Pick one theory and attach its factual basis |
| You repeat facts without saying why | The inferential bridge | Fact → element → why, one at a time |
| You conclude before tracing events | Sequence | Build the timeline, then change one event |
| You only ever see one side | Competing interpretation | State the strongest contrary reading before concluding |
| You do fine with labels, badly without | Cue dependence | A new setting with the topic headings removed |
| You do fine today, badly next week | Retention | Reconstruct from memory after a delay, not a re-read |
| You map it well but cannot write it | Organization and production | Order the paragraphs first; then write from the order |
A twenty-minute session that fits a real week
- Three minutes. Retrieve a topic you learned earlier — from memory, before opening the outline.
- Five minutes. Study one worked example that shows the decisions, not just the answer.
- Five minutes. Do a new example with support available.
- Four minutes. Compare a one-fact variation and say what moved.
- Three minutes. Read the feedback, name the error type, do the matching repair.
This allocation is a practical starting point, not a validated dosage. A five-minute version — one retrieval, one decision, one correction — is a real session, and interrupting it costs nothing. Return to material after a day, then three, then a week, then three weeks, adjusting from how it goes rather than from the schedule.