Skip to content
Need Help? Sign in
  • There are no suggestions because the search field is empty.

How to Research a Judge Using the Trellis Connector | Trellis Support

A step-by-step workflow for moving from a single data point to a litigation strategy — using Trellis judge analytics inside your AI platform.

Overview

The Trellis Connector brings Trellis's state trial court data directly into AI platforms like Claude or Chat GPT, letting attorneys research judges, analyze rulings, and develop litigation strategy without switching platforms or starting from scratch.

This article walks through a complete four-step research workflow using Claude as an example. The example involves the Hon. Gary Y. Tanaka of the Los Angeles County Superior Court and his approach to demurrers in labor and employment cases. The same workflow applies to any judge, motion type, or practice area covered by Trellis.

For a deeper look at the analytical principles behind this workflow, see the Trellis blog: No Assembly Required: Prompting Strategies for Judicial Research.

🔎 To get started with the connector, see Trellis Law MCP Connector.
🔎 For privacy information, see Privacy & Confidentiality When Using the Trellis MCP Connector.

What You Need

Step 1: Get a Baseline

Start with a direct request. Ask your AI platform to pull the judge's ruling history on the motion type you are researching, and tell it to use Trellis.

Example prompt:

How has the Hon. Gary Tanaka of the Los Angeles County Superior Court ruled on demurrers? Use Trellis.

The model returns the judge's grant rate and compares it to county and statewide averages. In this example, Tanaka sustains demurrers 55.5 percent of the time, slightly above both benchmarks.

This gives you a starting point. But a countywide average covers a wide range of dockets. The more meaningful comparison is how this judge performs against others handling similar cases — which is what Step 2 addresses.

Step 2: Compare Against a Relevant Benchmark

Rather than comparing to the broad county average, ask your AI platform to compare the judge to others handling similar cases. This tells you whether the judge's behavior is meaningfully different within the right peer group.

Example prompt:

How has the Hon. Gary Tanaka of the Los Angeles County Superior Court ruled on demurrers? Compare his approach to other judges handling similar cases in Los Angeles County. Use Trellis.

The model identifies other judges with comparable dockets and surfaces how Tanaka's grant rate compares to theirs specifically — not just to the county as a whole.

🔎 For analytics on how a specific judge rules on motions — win rates, timing, and tendencies — see Judge Research.

Step 3: Look at What the Numbers Are Not Showing You

Two judges can have identical grant rates while approaching a motion in completely different ways. Grant rates summarize outcomes. They don't explain what happens inside the courtroom to produce them.

Once you identify judges with similar outcomes, ask your AI platform to review actual rulings and identify differences in how they analyze the same type of motion.

Example prompt:

Tanaka and Hammock sustain demurrers at nearly identical rates. Review a sample of rulings from both judges in labor and employment cases. What is the primary focus of each judge's analysis? Provide your response in a chart that gives a clear snapshot of their judicial reasoning, supported with concrete examples.

In this example, the two judges do very different analytical work despite identical grant rates:

  • Judge Hammock organizes his rulings around legal obstacles to recovery — the defects in the claim that make it impossible to proceed as pled.
  • Judge Tanaka begins with procedural framework — whether the claim has been brought through the correct legal pathway.

Knowing which issues a judge prioritizes helps you understand what arguments are likely to matter in that courtroom — information that doesn't appear anywhere in the grant rate itself.

🔎 For tips on finding and researching judges in Smart Search, see Judge Research.

Step 4: Turn the Analysis Into a Decision Framework

Research becomes useful when it helps you make a specific decision. Once you understand how a judge reasons through a motion, ask your AI platform to connect that reasoning to the decision you are actually facing.

Example prompt:

Assume a defendant is evaluating whether to file a demurrer to a complaint in Tanaka's courtroom. Based on the grant rate of his demurrers and the reasoning reflected in his tentative rulings, what strategic considerations should inform the defendant's decision? Present your response in a chart.

This reframes the question. Instead of asking whether a demurrer will be sustained, the analysis helps you evaluate what happens if it is. Some pleading defects lead to routine amendments that simply teach the plaintiff how to revise. Others force a new statutory basis, a procedural prerequisite, or a material narrowing of the claim — outcomes with very different strategic value.

Filing a demurrer is not just a question of whether you will win the motion. It is a question of what winning actually accomplishes.

🔎 For the full analytical framework behind this workflow, see the Trellis blog: No Assembly Required: Prompting Strategies for Judicial Research.

Tips for Getting Better Results

  • Always specify the judge's full name and court. This ensures Trellis pulls data for the correct judge, especially in large counties where multiple judges share similar names.
  • Specify the case type when comparing judges. Demurrer rates in contract cases may differ significantly from those in labor and employment cases. Narrowing the comparison produces a more relevant benchmark.
  • Ask for a specific output format when it helps. Requesting a chart or decision matrix makes complex comparisons easier to scan. Ask for narrative explanation when you need the reasoning laid out.
  • Name what's missing from an answer before asking the follow-up. The clearest follow-up prompts identify the gap in the previous output and ask specifically for what would fill it.
  • Frame your final prompt around the decision you are making. Research only becomes useful when it informs what you do next. Keep your specific decision in view when you ask the last question in the sequence.

Related Articles