Prompt Pilotby Write.law
All prompts

Research

Work out how this judge decides this motion

Builds the research plan for the judge in front of you: standing orders first, then the written record, then the procedural habits that decide outcomes, with an explicit line between what is knowable and what is courthouse talk.

About 20 minintermediateLitigation

Your prompt5,985 characters

Still to fill in: The judge, The court, The motion

RoleYou are a litigator who learned that a standing order decides more motions than a judicial temperament does, and who has watched a colleague lose a motion because nobody read the chambers rules. You separate what is on the record from what is a pattern from what somebody said at a bar lunch, and you label each one. You never characterize a named judge's views from memory, because you do not have that knowledge and the guess is worse than the gap.What I needBuild the research plan for The judge in The court on a The motion, and tell me which three decisions about my filing it should change.InputsJudge: The judge Court: The court Motion and posture: The motion What I already know: What you already know What I can search: What you can searchHow to work this1. Start with what is published and binding, because it is the cheapest and the most decisive: standing orders, chambers rules, the court's local rules for The court, any general order governing this motion type, and the scheduling order already entered in my case. Name where each lives and what to look for in it. A The motion that violates a chambers rule is denied before its merits are read. 2. Then the written record. Name the categories to search and the search to run in What you can search: published opinions by this judge on this motion type, unpublished orders sitting on dockets, and the verbs the judge uses when granting versus denying. Tell me what to extract from each order, not just to read them. Name no case and no ruling yourself. 3. Then the procedural pattern, which predicts better than the substantive one and is easier to establish: whether this judge sets a hearing on a The motion at all, rules from the bench or takes it under submission, refers matters out, requires a joint submission or a certification first, tolerates over-length briefs, and how long a ruling typically takes. Say where each of those is observable on a docket. 4. Then the reputational sources, each with an honest label attached: bar-association evaluations and judicial analytics products by category, the clerk's office for scheduling practice, prior counsel who have appeared, and a colleague's recollection. Label each finding record, inference from a pattern, or hearsay about a person, and say which ones you would put in an email to a client and which you would not. 5. Draw the line explicitly. Say in your own words what this research can establish, what it can only suggest, and what it cannot reach. Do not attribute any view, disposition, temperament, or ruling to The judge from your own knowledge, and do not name a case as one of theirs. Anything you name at all is marked [UNVERIFIED - confirm it exists and is this judge]. 6. Convert the plan into the three decisions it should change about this filing: how long the brief is, what the first paragraph does, and whether to request oral argument. If the research will not change one of the three, say so.Ask me firstYou have the judge, the court, the motion, and what I already hold. Ask me these three, which none of that supplies, then stop and wait: 1. Am I the moving party or opposing, and would I rather this be decided quickly or decided well? Those pull in opposite directions on whether to ask for a hearing. 2. What is my relationship with this court: first appearance, regular, or coming in after something went badly here before? 3. How much of this is worth doing? A discovery motion and a dispositive motion justify very different amounts of docket reading, and I would rather you tell me the cheap version than build the thorough one nobody runs. Do not begin until I answer. If I tell you to proceed anyway, state each assumption at the top of your output and mark it [ASSUMPTION - verify].Output formatFour ordered blocks matching steps 1 through 4, each listing what to look for, where it lives, the search or docket action that finds it, and roughly how long it takes. Then a labeled findings frame showing how each item should be marked as record, inference, or hearsay. Then the explicit statement of what this research cannot reach. Then the three filing decisions, each with the finding that would change it and the change it would produce. Close with one line naming the two of my answers that most changed this plan, and what you would have told me to do without them. If an answer changed nothing, say so, because it means the question was wasted.Never do this- If the plan would work for any judge on any motion, it is too generic. Everything has to engage with what a The motion in The court actually turns on. - No hedging filler. "This judge may be receptive" is not a finding and is not researchable. Say what to look for and where. Do not tell me to consult an attorney; I am the attorney. - Never state, imply, or guess at any ruling, opinion, view, temperament, grant rate, or preference of The judge, and never name a case as one of theirs. You do not have that knowledge. Anything you name is marked [UNVERIFIED - confirm it exists and is this judge]. - Where you cannot tell whether a practice is a rule, a habit, or a story, say you do not know and say which source would settle it. Do not smooth over the gap with a confident characterization of a real person. - Do not pad. If the standing order settles the question, say so and stop. Length is not value.Before you answer- Did I name any case, ruling, statistic, or view and attribute it to this judge? - Is every item labeled record, inference, or hearsay? - Could a first-year run every search and docket action exactly as written? - Did I say which findings belong in a client email and which do not? - Would this plan be useless for a different motion in a different court? It should be.

The run walks turn one, the pressure test, the follow-up, and a check on what came back. The Cockpit adds driver's-seat tunes. Your values come with either one.

2

Pressure-test it

Makes the AI switch hats and attack its own answer.

Your adversary has done this same research and reached the opposite conclusion about how this judge handles a The motion. Work through the findings as the adversary who did it: which of my conclusions rests on a single order, which on a pattern thin enough to be coincidence, and which on something a person said about another person at a bar function. Then tell me which finding I should stop relying on, and what the honest version of my filing plan looks like once it is gone.
3

Go deeper

Pushes the work further once the basics are right.

Research about a judge is worth only the filing decisions it changes. Write the one-page filing plan for this motion: the length and why, what the first paragraph has to do, whether to request argument and the sentence that requests it, the standing-order requirements most filers miss in this court, and the single procedural practice that changes the structure of the brief rather than its tone.
4Check what came backPaste the answer here and work a checklist against this prompt's own rules.

Before you run it

What to gather first

  • The judge and the court, exactly as the docket names them
  • The motion you are filing and its posture
  • Anything you already know: a prior appearance, a standing order you have read
  • Which research tools and analytics you actually have
  • The deadline, since some of this takes a day and some takes an hour

Watch for

  • The model has no knowledge of any individual judge and will invent rulings, statistics, and preferences that sound authoritative. Treat every name or figure it produces as fabricated.
  • Judicial analytics products report what can be coded from dockets, which is a thin slice of what decides motions. A grant rate across all movants tells you little about your motion.
  • Standing orders and chambers rules change without notice and are often not where you expect. Read the current version from the court's own site before you rely on anything about form or procedure.
  • Anything you write about a judge can be produced, forwarded, or read aloud. Keep characterizations of a person out of email and out of the file.
  • Researching a judge is legitimate; contacting chambers about the merits is not. Confine any call to the clerk to scheduling and form.

What comes back

Four ordered blocks (binding published materials, the written record, the procedural pattern, the reputational sources), each with what to look for, where it lives, the search or docket action, and the time it takes. Then a labeling frame marking every finding as record, inference, or hearsay, an explicit statement of what the research cannot reach, the three filing decisions it should change, and the answer-consumption line.

See an example of what you’ll get
*(After I answered: I am the moving party, I would rather have it decided quickly than perfectly, this is my first appearance before her, and it is a discovery motion so I want the cheap version.)* 1. Published and binding. Do this first: it is an hour and it decides the most. | Look for | Where | How | Time | |---|---|---|---| | This judge's standing order for civil cases | The court's own website, judge's page, not a database | Read it end to end, twice for the discovery section | 20 min | | Chambers rules or civil standing order on discovery disputes specifically | Same page; often a separate document from the general standing order | Note whether a joint letter, a pre-motion conference, or a certification comes before any motion, and the page limit for it | 15 min | | Local rules of this court on discovery motions | Court website | Note the meet-and-confer certification requirement and its form | 10 min | | The scheduling order in my own case | The docket | Confirm the discovery cutoff and whether motions must be heard rather than merely filed before it | 5 min | You have already told me a joint letter is required before any discovery motion. If that is right, it changes the deliverable entirely: what you are writing is a letter section, not a motion, and everything below about brief length is moot until you confirm it. [UNVERIFIED - confirm the current standing order; it is the single highest-value item here and the one most often out of date in a colleague's memory.] 2. The written record. Two hours, and skip it if the joint-letter procedure holds. Search categories rather than names. In your database, run a search restricted to this court and this judge for written discovery rulings, sorted newest first, and read the five most recent. Extract three things from each: the procedural route the dispute took to get to her, whether she resolved it on the papers, and the verb she uses in the disposition. On the docket side, discovery orders are frequently unpublished, so a database search alone will miss most of them; pull two or three docket sheets of closed cases before her and read the discovery entries directly. I cannot name a single one of these orders, and any citation you see me produce would be invented. 3. The procedural pattern. Observable from dockets, and more predictive than anything in block 2. Whether a hearing is set at all on a discovery dispute; whether the entry says "submitted without argument"; the interval between filing and order; whether she orders the parties back to meet and confer instead of ruling. All four are visible from the docket text of three closed cases and cost about forty minutes. That interval number is the one that matters to you, because you said speed is what you want. 4. Reputational sources, labeled. | Source | What it gives you | Label | |---|---|---| | The standing order | The procedure, bindingly | Record | | Docket entries across three cases | The pattern of how disputes resolve | Inference from a pattern | | The courtroom deputy, on scheduling and form only | When she hears discovery matters, what format she wants | Record, if you confine the call to logistics | | Your partner's two appearances in 2024 | A recollection of one person about another | Hearsay about a person. Useful for orientation, and it does not go in an email to a client | | A judicial analytics product, if you had one | Coded outcomes across all movants | Inference at best, and you told me you do not have one | What this research cannot reach. It cannot tell you how this judge will view your specific dispute, what she thinks of proportionality arguments, or whether she is inclined toward or against your side. Nothing in a docket establishes that, and any source that claims to is selling a characterization of a person. Treat the procedural findings as reliable and the substantive ones as orientation. The three decisions. - Length. Governed entirely by the standing order, not by preference. If the joint-letter procedure holds, you have a hard page allocation and the decision is made for you. - What the first paragraph does. If the docket pattern shows she orders parties back to meet and confer, the first paragraph is not about the documents. It is about what the two meet-and-confer rounds already produced and why a third would produce nothing. - Whether to request argument. You want speed. If the pattern shows she resolves discovery disputes on the papers in under three weeks, requesting a hearing costs you time and buys nothing. If the pattern shows hearings, asking for one costs nothing and may shorten the interval. What your answers changed. That you want this decided quickly rather than perfectly is why block 3 is where the real work is and block 2 is marked skippable. On a dispositive motion the order reverses: you read the written record for how she reasons and the docket pattern becomes background. That this is your first appearance is why the courtroom deputy is in the plan at all and why your partner's recollection is labeled the way it is; if you were a regular here, the deputy call is unnecessary and the recollection is worth more. Your answer that you are the moving party changed the argument-request analysis at the margin only, and I could have built the same plan without it.
Why this prompt is built the way it is
## Framework 1. **Published and binding first.** Standing orders, chambers rules, local rules, the scheduling order. These decide more motions than preferences do. 2. **Then the written record.** Opinions on this motion type, unpublished orders on the docket, the language used when granting versus denying. 3. **Then the procedural pattern,** which usually predicts better than the substantive one: hearings or papers, bench rulings or submission, referral to a magistrate, tolerance for length. 4. **Then the reputational sources, each labeled** as record, inference, or hearsay about a person. 5. **Draw the line out loud** between what is knowable and what is rumor. Never attribute a view to a named judge from memory. 6. **Convert findings into three filing decisions:** length, what to lead with, and whether to ask for argument.