The State Transition Model in DILR

Nine minutes into a DILR set, an aspirant stops and stares at a half-filled grid. Not because they are stuck on a deduction. Because they cannot remember whether the thing in row three was something they proved or something they were trying out.

That moment costs more DILR marks than any reasoning failure. The logic was fine. What broke was bookkeeping: the set had moved through six or seven states, and only the latest one existed anywhere, in a head already busy holding the next deduction.

A DILR set is not a puzzle you solve. It is a system that changes state every time you learn something, and the solvers who finish are the ones who track the state instead of trying to remember it.

Why DILR Sets Punish Memory

The standard approach is to draw a grid and fill it in. That works while there are few facts. It stops working at the point where the number of things you know exceeds what you can hold, which for most people arrives around the fourth or fifth deduction.

The moment it breaks is fairly consistent across aspirants:

• Deductions one to three sit comfortably in your head alongside the grid.
• Around the fourth, you start re-checking things you already established.
• By the sixth, you are re-reading clues to confirm what the grid already says.
• Past that, you are no longer deducing, you are auditing, and the clock does not care about the difference.

What makes it worse is that a grid records conclusions but not their status. A letter written in a cell looks identical whether it was deduced, assumed for a case, or pencilled in as a guess. Ten minutes later you cannot tell them apart, and the only safe response is to redo work you have already done.

The specific ways state gets lost are predictable:

• Assumptions become facts. A trial value written the same way as a proven one, and later treated as proven.
• Eliminations vanish. You ruled out three options for a slot, wrote nothing down, and have to rule them out again.
• Used clues get re-read. Without a record of which constraints are spent, you cycle through all of them repeatedly.
• Progress becomes invisible. You cannot tell whether the last two minutes narrowed anything, so you cannot tell whether to continue.

The State Ledger: Three Columns, Updated After Every Deduction

The ledger sits beside the grid, not instead of it. The grid holds the arrangement; the ledger holds what you know about it and how firmly.

Fixed is the smallest column and the most valuable, because it is the only thing you never have to re-derive:

• Write it in a different notation from everything else, so it is visually unmistakable.
• Add the clue number that fixed it, so you can check it later without redoing the reasoning.
• Never write a case assumption here, no matter how confident the case feels.

Narrowed is the largest column and the one most solvers skip entirely, which is why they repeat eliminations:

• Record the surviving options, not the eliminated ones. Shorter to write and faster to read.
• When an option list reaches two, mark it. Two-option slots are where the next deduction usually comes from.
• Update in place rather than rewriting. A ledger you rewrite is a ledger you stop maintaining.

Untouched is the column nobody keeps and the one that changes how you search:

• List every entity and slot at the start. Cross each off as any clue touches it.
• What remains is where new information has to come from, so it tells you which clue to re-read.
• If the column is empty and the set is unresolved, the answer is in combining clues rather than finding a new one.

What a State Change Actually Looks Like

Thinking in transitions rather than additions changes what counts as progress. A deduction that adds a fact but narrows nothing has not moved the set forward.

The third row is the one that quietly eats time. Re-reading a clue that has already been fully used feels like work and produces nothing, and without a ledger there is no way to know you have done it before.

How to Keep the Ledger Cheap

A ledger that costs thirty seconds per deduction will be abandoned by minute four. These constraints keep it at ten.

• Symbols, not sentences. A circled letter for fixed, a small list for narrowed, a tick for touched. Nothing gets written twice.
• One line per entity. Not per clue. Clues are inputs; entities are what has state.
• Never erase. Strike through instead. An erased elimination is an elimination you will redo.
• Number your clues and tick them as they are spent, so a spent clue is never re-read.
• Update immediately. A deduction held in your head for two more steps is a deduction you will misremember the status of.

The same discipline is what makes an early re-read productive rather than panicked, which is covered in why every impossible DILR set is actually waiting for one observation.

Common Mistakes That Lose the State

Related habits that cost sets:

• Grid only, no ledger. Recording arrangements while discarding everything about their status.
• Erasing eliminations. Cleaning up a working grid and destroying the reasoning that produced it.
• Re-reading all clues each round. Cycling through spent constraints because none are marked as used.
• Tracking only what is fixed. Ignoring the narrowed column, which is where nearly every next step lives.
• Never listing the untouched. Searching randomly for the next deduction instead of looking where information has not yet reached.

A Practice Drill for Tracking State

This drill separates bookkeeping from reasoning so you can see which one is failing.

1. Take five DILR sets you have already solved, with the solutions available.
2. Re-solve each one, but keep the three-column ledger alongside the grid, updating after every single deduction.
3. Whenever you finish, count how many times you re-read a clue you had already used. Write the number down.
4. Repeat on five fresh sets a week later and compare the re-read counts.

What aspirants typically find:

• Re-read counts start high, often six or more per set, and halve within two sessions.
• Time to first fixed value barely changes, but time from there to completion drops sharply.
• Sets previously abandoned around minute nine start finishing, because that is exactly when memory used to fail.
• The untouched column, once kept, points at the decisive clue more often than intuition does.

The Bottom Line

DILR does not test how much you can hold in your head. It tests whether you built something you can trust when your head is full. The grid records where things are; the ledger records what you know, and only one of those survives minute nine.

The State Ledger, Recap

• Fixed: certain and permanent. Never write an assumption here.
• Narrowed: fewer options than before, with the surviving list recorded.
• Untouched: not yet constrained by any clue. This is your search list.

Build the Habit on Timed CAT DILR

Bookkeeping is the first casualty of time pressure, so it has to be rehearsed under it. Work through CAT previous year questions in timed blocks, or sit full CAT mock tests and past papers so the ledger survives real section fatigue. More DILR approaches sit in the CAT DILR blog archive, and if sets keep collapsing late it is worth having your strategy reviewed honestly.

The Optima Learn Editorial Team builds CAT preparation content from exam-pattern analysis and Optima Learn's adaptive practice data. This guide is part of our CAT DILR preparation series.