The Problem with "Tough but Fair"
I have been appearing in New York State Supreme Court for more than 28 years. Five boroughs plus Nassau, Suffolk, Orange, Putnam, and Rockland. Thousands of appearances across dozens of parts. And I can tell you that the single most common question attorneys ask when they get a case assigned to an unfamiliar judge is some version of: "What's this judge like?"
The answer they get back is almost always useless. "Tough but fair." "She runs a tight ship." "He likes to settle." These are not insights. They are personality descriptions. They tell you nothing about how the judge actually decides motions, what arguments she finds persuasive, or what procedural missteps will get you in trouble in that specific courtroom.
I know this because I have built my practice around knowing the difference. When I cover a conference in Part 96 in Kings County, I know that Justice Sheares runs one of the slowest calendars in Brooklyn. Cases are still being called after 12:45 PM. If you send a covering attorney there expecting a 90-minute morning appearance, you are going to have a problem. That is not something you learn from "tough but fair." That is something you learn from being in the room.
What You Are Missing
Every judge has patterns. Not reputation. Not vibes. Actual patterns in how they handle cases, and those patterns are documented in their own orders and decisions.
In Queens, Justice Dufficy in Part 35 requires hard copies of motion papers. If you file electronically and assume that is enough, you are starting behind. Justice Livote in Part 33 decides motions on submission with no oral argument unless the Court requests it. If you are preparing a ten-minute oral argument for Part 33, you are wasting your time. These are specific, actionable facts that change how you prepare for an appearance.
Some judges compress discovery timelines in ways that catch attorneys off guard. Queens County runs 60-day deadlines in certain case types. If you come in expecting the standard schedule, you are already late. Some judges require Affirmations of Good Faith before they will hear a discovery dispute. Others will sanction on a first violation. The difference matters, and it is all sitting in publicly available court records.
The tools that aggregate this kind of data, Lex Machina, Trellis, Pre/Dicta, are built for AmLaw 200 litigation departments. If you are a solo or a five-attorney shop already paying $270/month for Lexis and $96 for Westlaw Essentials, another $900 for analytics is not in the budget. So you go in on anecdote. On what a colleague told you three years ago about a judge who may have changed their entire approach since then.
What I Actually Do
I maintain intelligence on individual judges across every county where I make appearances. Not from a database product. From the public record, from nycourts.gov published part rules, and from my own appearance reports going back years.
Here is an example. When Queens County reshuffled its bench in early 2026, I had a sub-attorney on the ground who documented every change: Justice Sampson retired out of Part 31, Justice Lin moved from Long Island City to Jamaica at Part 24, Justice Maldonado Cruz moved to Part 8G. I verified what I could against nycourts.gov and flagged what was still unconfirmed. That is how I learned that the judge listed as "Rory Lancman" in some directories is actually a different person from "Mojgan Cohanim Lancman," who is the sitting justice.
That kind of detail matters when you are preparing for an appearance. It is the difference between walking in prepared and walking in with stale information.
I do the same work in every county. In Kings County, I can tell you which parts run slow calendars and which move quickly. In the Bronx, I know the CCP procedures, the City parts, the med-mal rotation. In Nassau, I have documented which judges insist that discovery proceed regardless of pending motions (Justice Grimaldi, for one) and which judges will grant a stay on an OSC (Justice Genovesi). In Suffolk, I know the courthouse at 1 Court Street in Riverhead and the local procedures that trip up attorneys who are used to practicing in the city.
How You Can Build This Yourself
You do not need my 28 years of experience to start building judge intelligence for your own cases. You need two things: NYSCEF and a local AI tool that reads files on your computer.
I want to be specific about what I mean by "local AI tool." I do not mean ChatGPT or any browser-based chatbot. Those tools respond from training data. If they do not know something, they make it up. That is how we got Mata v. Avianca and more than 600 cases with fabricated citations. Lawyers asked a chatbot to find supporting case law and trusted what came back without checking.
A local tool like Claude Code runs on your own machine and reads the actual files in your folders. It does not generate case law from memory. It reads the documents you give it. That is a fundamental difference.
Here is the workflow:
Pull the judge's recent decisions from NYSCEF. Go to iapps.courts.state.ny.us/nyscef and search by judge name. Download the last 30 to 50 orders and decisions as PDFs. The interface is not pretty, but the documents are there. For federal cases, PACER has the same material.
Put everything in one folder. The judge's decisions, your complaint, the motion you are opposing or drafting, your exhibits. Everything the tool needs to see.
Point the tool at the folder and ask questions about the judge. Not "find me cases that support my argument." Ask questions like:
"In these 40 orders, how often does this judge grant motions to compel? What does she require before she will hear a discovery dispute?"
"When this judge denies summary judgment, what factual issues does she focus on?"
"Does this judge ever raise issues sua sponte? What are they?"
"How does this judge handle adjournment requests? Is there a pattern in what gets granted and what gets denied?"
Check the answers against the actual documents. This is the step most people skip, and it is the most important one. Read what the tool tells you, then verify it against the orders. If it gets something wrong, correct it. The tool updates its understanding within that session. This is not prompting. This is the same review process you would use with a junior associate doing a research memo.
Once you have a reliable profile of the judge, you can ask the tool to help you structure your brief or opposition in a way that reflects how this specific judge reads arguments. What issues does she spend the most ink on? What procedural requirements does she enforce strictly? What arguments has she rejected repeatedly? That is how you move from a form brief to a brief written for the judge who will actually read it.
Why This Matters More in State Court
The federal analytics tools have started to do this work on the PACER side. If you practice in SDNY or EDNY, you can buy access to some of this data. But in New York State Supreme Court, almost nobody is doing systematic judge analytics. The data is in NYSCEF. It has been there for years. Most practitioners are not using it because the interface is difficult and the volume of documents is large.
That is exactly why a local AI tool changes things. Reading 40 PDFs and extracting patterns from them used to be a full week of associate time. Now it takes an afternoon. The barrier was never access to the data. It was the cost of reading it at scale. That cost just dropped to nearly zero.
If you practice in New York State Supreme Court and you are not building judge profiles from NYSCEF, you are leaving an advantage on the table that your opponents may already be picking up.
A Note on AI and Legal Ethics
Nothing in this article suggests using AI to generate case citations or legal authority. The workflow described here uses AI to read and analyze documents you already have, the same way you would use a research associate. There is a clear difference between asking a tool to invent case law and asking it to read 40 actual orders in a folder on your computer. Verify everything. Read the underlying documents. The tool is as good as your supervision of it.
This Is What We Do Every Day
At the Law Office of Frederic R. Abramson, judge intelligence is not a theory. It is the foundation of every appearance we make. We maintain individual judge profiles across all five New York City boroughs, Nassau, Suffolk, Orange, Putnam, and Rockland counties, built from more than 28 years of courtroom experience and verified against public records.
When you hire us to cover a conference, a motion, or a deposition, we walk in knowing how that specific judge runs their courtroom. Not because someone told us. Because we have been there, and we have done the work.
If you need per diem coverage from local counsel who actually prepares for the part they are walking into, we should talk.