This Is the Sixth Time You've Started Over

Sunday, half nine, and you're not thinking about the chicken

The shop is done. The app is reinstalled. You've told yourself Monday.

And you're sitting on the edge of the bed thinking about your sister-in-law in August, and about the person at work who said something nice in month four last time and then, around month nine, stopped saying anything at all — which was worse.

You know how to lose three stone. You've done it five times. That's the problem. You're not afraid of the diet. You're afraid of the part after it, in public, again.

Nearly everything you can buy stops writing at week twelve: the plan, the first thirty days, the shopping list, the before photo. Months four to nine barely exist, which is remarkable, because that's where all five of yours died. The advice that does exist for that stretch is stay consistent and make it a lifestyle — restatements of your problem at the emotional temperature of a motivational poster.

So don't buy a sixth beginning. You own five. Spend this week reading them instead.

Get a piece of paper. One column per attempt, five questions down the side. It takes about forty minutes and it's the only thing in this article.

1. What month did it stop, and what else was happening?

The rest of your life, not the diet. A job change, a house move, a bad winter, someone ill, a relationship ending.

Nobody has ever asked you this. Everyone asks how you did it. Nobody asks how it stopped, because stopping gets treated as a character flaw rather than an event with a cause — and it had a cause, probably the same one, five times over.

If something shows up in four of the five columns, that's the actual subject of this attempt, and it may have almost nothing to do with food.

2. What was the first thing to go?

Tracking, cooking, weighing, walking. They don't fall at once. One goes first and it's usually the same one, and it's usually the one that costs the most time on the worst day.

That's your early warning. Not the scale — the scale tells you weeks later. The tracking going quiet on a Wednesday tells you now.

3. Were you still losing when it stopped, or had it flattened out?

Different failures entirely. Quitting during a plateau is a story about expectations. Quitting while it's still working is a story about something else, and the something else is usually in column one.

4. What did people say to you in the two weeks before?

Compliments included. Especially compliments.

This is the part that's keeping you on the edge of the bed at half nine, and no article about calories will touch it. You've been publicly congratulated and then publicly, silently un-congratulated, five times. A sixth round of that is a genuinely unpleasant thing to sign up for. It isn't vanity and it isn't fragility.

So don't sign up for it. Nothing here requires an audience. You can decide tonight that precisely one person knows, and that it's somebody who has never once commented on your body. You can decide nobody knows. You can decide that when someone asks what you're doing differently, the answer is that you're sleeping better, followed by a change of subject. None of that is dishonest. Your weight isn't public information and you never agreed to run a commentary on it.

The cost is real, though. Some of those five attempts were partly powered by being seen, and this one won't be. It'll feel flatter. Plan for flat.

5. What were you eating when you weren't following the plan?

Write down where you were standing and who else was in the room. Alone or with people. Evening or all day.

The worst hour is the one right after — you've eaten everything in the house, you're in the kitchen at eleven, and the conversation you need is now, not Thursday, and not with anyone whose face is going to change. TrueTalk has an AI persona called Dr. Maya Wellness, listed under weight management, nutrition planning and exercise routines. Not a doctor and not a dietitian, awake at eleven when neither of those is, free the first time.

Then plan the month it ends in

Before you decide a single thing about breakfast, decide what happens in the bad month.

Say the forensic turns up that every attempt ended within a fortnight of something going wrong at work. Then the question isn't what do I eat, it's what does this look like during a fortnight of coming home at nine. Answer it now, in writing, while you're calm. What's the minimum version? Which things do you keep, and which do you drop deliberately, in advance, without it counting as stopping?

Pick a smaller target while you're there. Three stone needs a good year, and a good year is exactly what your history says you don't reliably get. A stone, or twelve weeks of a specific behaviour rather than a weight at all, has the property that a bad month doesn't invalidate it. A goal that can absorb a bad month is the only kind your own evidence says survives one.

I'll be straight about what that costs: smaller target, smaller and slower visible change, and less of the fuel that came from people noticing. You're trading that away on purpose, for the thing not ending.

The bit that needs someone with a licence

A long history of losing and regaining the same weight is worth a GP appointment, and so is anything that looks like a binge pattern — eating a large amount fast, alone, past comfort, with shame afterwards. That's a recognised clinical picture with actual treatment behind it, and it doesn't respond to a better meal plan. A referral to a registered dietitian is a different and more useful thing than another programme, and in a lot of places you can also self-refer for talking therapy.

Worth saying too: whether repeated weight cycling itself does long-term physical harm is genuinely argued about among researchers. Anyone who tells you that question is settled, in either direction, is going beyond what the evidence supports. Don't let it be used to frighten you into starting on Monday.

About Monday

Nothing about a Monday makes weight come off. It's a convention, and one reason tonight feels so heavy is that you've handed a normal weekday the job of delivering a verdict on you.

Do the five columns instead. You already have the data. You paid for it five times.