What Should I Eat

Why this question is so hard

Deciding what to eat is difficult in a way that is disproportionate to its importance. The options are usually all acceptable, which is exactly what makes it hard — when every choice is fine, there is no information left to choose with, and the decision stalls.

Add a second person and it stalls harder. Both are trying to accommodate the other, neither wants to impose, and the conversation converges on “I don’t mind, what do you want?” repeated until somebody gives up. The problem is not a lack of options; it is that nobody wants to own the decision.

A draw solves the ownership problem. Nobody chose, so nobody is responsible, and the meal happens.

The list is where the thinking goes

All the judgement belongs in the list. Put in what is genuinely available — what is open, what is in the fridge, what fits tonight’s budget and the time you have. A list containing something that takes ninety minutes to cook is not useful at eight in the evening.

Editing the list is also the part that resolves most of the difficulty. Writing down six options frequently makes it obvious that two of them are not real candidates, and occasionally makes it obvious that one of them is what you wanted all along. That is a legitimate outcome — you can stop before drawing.

Weighting for things you want often but not always

Weights let you express a real preference without turning it into a decision. “Pizza | 3” against everything else at 1 means pizza wins more often but the rest are genuinely in play. That is closer to how preferences actually work than either picking pizza every time or pretending all options are equal.

It also keeps the possibility of a pleasant surprise, which removing options entirely does not.

Reading your own reaction

The most useful moment is the second after the result appears. Disappointment means you wanted the other thing — so have the other thing, and note that the draw found the answer faster than the discussion would have. Relief means the option was better than it looked on the list.

Used this way the tool is less a decision-maker than a way of surfacing a preference you had not articulated. There is no obligation to obey it, and treating the result as binding misses most of the value.

Planning several meals

Drawing several options at once with repeats off produces a varied week rather than a single choice. Done at the start of the week it feeds directly into a shopping list, and it removes the daily deliberation entirely — the decision was made once, impartially, and the rest of the week just executes it.

This also tends to produce more variety than deciding day by day, because same-day decisions gravitate toward whatever is easiest, which is usually the same handful of things.

What this tool does not do

It has no view on nutrition, cost or quality, does not know what is near you, and is not connected to any restaurant, delivery service or brand. The preset lists are ordinary generic options provided as a starting point to edit. Every option is equally likely unless you weight it, and nothing about the draw reflects any judgement about the food itself.

Privacy

Your list is processed in your browser. Nothing is uploaded, stored or placed in the page link.

Deciding for a group

The tool helps most where the difficulty is social rather than culinary. When several people each have a mild preference and nobody wants to override anyone, the draw resolves it without anybody having to advocate for their choice.

Building the list together is part of that. Each person adds one or two options they would genuinely be happy with, which quietly filters out anything somebody actively dislikes before the draw happens. The result is then acceptable to everyone by construction rather than by negotiation.

Dietary requirements belong in that filtering step rather than in the draw. Anything that will not work for somebody at the table should not be on the list at all, because a draw treats every line as equally acceptable.