What to cook with what I have tonight

Most evenings that end in a takeaway do not fail at the cooking stage. They fail earlier, at the fridge door, in the thirty seconds where you look at a shelf of unrelated ingredients and nothing assembles itself into a meal. That is a decision problem. Cooking skill does not fix it, and neither does another recipe app with a hundred thousand recipes. Fewer options, in a sensible order, does.

Start from what is leaving soonest

Do not start with "what do I feel like eating". That question is open at both ends and it is the one you have already failed to answer. Start with a much narrower one: what in this fridge has the least time left?

Usually the answer is a small, specific list — half a bunch of coriander, the mushrooms, an opened cream. Now your problem has become "what can I make with mushrooms tonight", which is a question with obvious answers, rather than "what is dinner", which is a question with infinite ones. Constraint is not a consolation prize here. It is the mechanism.

There is a quiet second benefit. Cooking the thing that was about to be wasted feels good in a way that has nothing to do with virtue. It is the pleasure of a loose end being tied off.

Sort by how far you are from the finish line

Then split what you find into three piles. This is the whole method.

1. Can cook right now

Everything needed is in the house. On a tired Tuesday this is the only pile that matters, and it should be the first thing you see. The value of this pile is not that the food is better; it is that choosing from it costs you nothing beyond choosing.

2. Missing exactly one thing

This pile is more useful than it sounds, because "buy one item on the way home" is a completely different act from "do a shop". If you look at it before you leave work rather than after you get home, it quietly doubles what you can make.

3. Missing several things

Not tonight. Worth a glance for the weekend, worth ignoring right now. Most recipe apps show you almost exclusively this pile, which is exactly why they do not help at 6pm.

Keep a small set of shapes, not a big set of recipes

People who cook fluidly from an unplanned fridge are rarely recalling recipes. They are filling in a handful of familiar templates:

• Something fried with rice. One protein, two vegetables, an aromatic, a sauce. Absorbs almost anything.
• A soup or a stew. The most forgiving container ever invented for slightly tired vegetables.
• Pasta with one dominant idea. Garlic and oil, tomato, or cream — then whatever you have, added to it.
• Eggs as the main event. Omelette, fried rice, shakshuka, savoury custard. Fast, and almost always possible.
• Roast a tray of it. Anything solid, hot oven, salt. Requires no attention while it works.

Five shapes cover most weeknights. Ingredients change; the shape does not. This is why "what can I make" gets easier over years while your recipe collection stays roughly the same size.

Two small habits that do most of the work

The first is seeing the fridge without opening it. Standing in front of an open door is the worst possible place to think, because you are scanning, cold air is escaping, and you are under mild time pressure. Any list you can look at from the sofa, or from the office at five o'clock, is better — even one written on the back of an envelope.

The second is deciding earlier than you need to. The difference between a good weeknight dinner and a takeaway is very often a decision made at 5pm instead of 7pm. Not more planning, not a meal-plan spreadsheet — just moving the same two-minute decision forward by two hours, so it happens while you still have options.

And on the nights it does not work, order the takeaway. This is dinner, not an exam. The mushrooms will still be there tomorrow, and tomorrow they will be at the top of the list.