Most people slip many times before a quit lasts, and what they see and hear after a slip matters

Most people slip, many times

Ask people who quit smoking how many tries it took and they give a small number. In an Ontario study, smokers who had stayed off for a year remembered 6.3 attempts on average¹. That number only counts the people who made it. Everyone still trying is missing from it, and so are the attempts people forget.

Chaiton and colleagues used the Ontario Tobacco Survey, which interviewed adults about every 6 months for up to 3 years between 2005 and 2011¹. 1,277 of them made 3,362 attempts to quit in that time¹. Only 170 of those attempts lasted a year, 5.1%¹. If every attempt had the same chance, that works out to 19.6 attempts for each success¹.

The chance was not the same. The first attempt the study saw lasted a year 11% of the time, and the second 5%¹. Following every attempt person by person gave the authors' best estimate, 29.6 attempts before one lasted a year, with a 95% interval of 27.6 to 31.6¹. They chose it as resting on the best set of assumptions and say it likely still counts too few. Adding the attempts people recalled from before the study pushed it to 142, which the authors call a probable overestimate¹.

This is smoking, and the counts are self-reported, with no test to confirm who had quit¹. It is also an average across a population. The authors say it cannot tell any one person how many tries they will need¹. What it does show is that a failed attempt is the normal case, by a wide margin. Someone who has tried to quit porn ten times and is still trying is not unusual. On these numbers, they are early.

What a first slip looks like

If most attempts end, the next question is how they end. With one slip, or with a slide? Kirchner, Shiffman and Wileyto followed 203 smokers in a nicotine patch trial at a quitting clinic, run from 1997 to 2000, who slipped at least once in the first 6 weeks after quitting². Each one carried a handheld computer and recorded how they felt at every slip, so the answers came in the moment and not weeks later.

Most slips were small. 79% were one cigarette or less². Most of these smokers also did not return to daily smoking. 111 of the 203, 55%, slipped but never went back to daily smoking in those 6 weeks, though they slipped about 5 times each on average². 64, or 32%, went back to daily smoking, meaning 3 or more days in a row, but did not relapse fully². 28, or 14%, relapsed fully, set at 5 or more cigarettes a day for 3 days in a row².

Guilt after the first slip did not decide it

The study set out to test the idea that a first slip does its damage through the feelings that follow. You break the rule, feel guilty, decide you have failed, and give up. The numbers did not bear that out. How guilty people felt after the first slip, how much they blamed themselves and how confident they were did not predict who relapsed². Confidence after the first slip did predict a faster return to daily smoking².

Across the whole run of slips, confidence mattered. Each 1-point fall in confidence after a slip, on a 0 to 10 scale, raised the rate of a further slip by 9%². Blame worked the other way from what you might guess. On its own it did not predict the next slip, but when a slip came after a longer stretch without smoking, people who blamed themselves more went longer before the next one². Guilt on its own did not predict the next slip either, though it sped the next slip for people on the nicotine patch and slowed it for those on placebo². So across the whole group, the danger was not feeling bad about a slip. It was losing faith, slip by slip, that quitting could work.

None of this makes an early slip harmless. A separate study of 1,190 smokers by Deiches and colleagues found that those who slipped in the first 8 weeks were less often off cigarettes at 6 months, 21.6% against 70.7%³. Kirchner and colleagues also cite older studies in which 85% to 95% of smokers who slip eventually relapse, and their own tracking stopped at 6 weeks after quitting².

Read together, the two studies say a slip is a warning sign that still does not settle the outcome by itself. What the slips do to a person's belief that they can stop looks like the part that counts². The smokers here were heavy smokers who sought treatment, mostly white and better educated than average². The pattern may not hold for everyone.

A broken streak on screen

If falling confidence is the risk, then what a person sees right after a slip matters. A counter that drops back to zero is about the loudest thing an app can show. Silverman and Barasch tested what a broken streak does to the choice to keep going⁴.

In one of their experiments, 602 adults recruited online answered Portuguese questions in a mock app modelled on Duolingo⁴. An intact streak was four in a row. A broken one was three, then a message that the fourth question already had enough testers. Nobody had failed. The app broke the streak itself⁴. Half saw their run marked in an in-app log, and the other half answered in a plain survey with no log.

With the log on screen, 92% of those with an intact run chose to keep going⁴. With a broken run, 45% did⁴. Without the log, 65% and 61% kept going, a gap the authors did not test on its own⁴. The break was identical in both groups. What changed was whether they saw it in the log inside the app or answered in a plain survey⁴.

Seven studies found the same pattern⁴. In one, people whose streak broke through their own failure kept going least, 29% against 53% for an intact streak⁴. That is the case closest to a lapse. A field study of 980 employees in a walking challenge pointed the same way, but it could not rule out that those who missed a day were already losing interest⁴.

The limits are real. These were paid online volunteers, studied between 2017 and 2021, doing a few questions, games or exercises in mock apps, with streaks of 3 or 4⁴. None of it involved days of recovery. Still, if losing a run of four questions in a made-up app halved how many kept going, we would not bet on a lost count of weeks clean doing less.

What to say after a failure

Hiding the count is one half. The other half is what a person hears after the slip. Breines and Chen gave 86 university students a 10 item vocabulary test hard enough that they averaged 4 out of 10⁵. Before a retest, they could study the words for as long as they liked. A short statement added to the instructions changed by group.

Students given no message studied for 203 seconds on average⁵. Those given a line meant to lift their self-esteem studied for 230⁵. Those given a line of self-compassion studied for 306⁵. That line began "If you had difficulty with the test you just took, you're not alone"⁵.

In another experiment, 91 students recalled something wrong they had done⁵. Those who wrote to themselves with compassion were more motivated to make amends and not repeat it, 4.97 out of 7 against 4.11 for self-esteem⁵. That fits with the smokers. Blame did not sink them and lost confidence did. A kind line keeps someone trying without pretending the failure never happened.

This was a lab study with groups of 27 to 30 students at one university⁵. Only the gap with no message was significant. Self-compassion against self-esteem was marginal⁵. Retest scores did not differ significantly by group, though students who studied longer scored higher⁵. The kind word bought effort, not a significant gain in scores, and the effort was counted in minutes.

How they found it

• The Ontario Tobacco Survey interviewed adults about every 6 months for up to 3 years. 1,277 people made 3,362 attempts to quit. The first attempt the study saw lasted a year 11% of the time, the second 5%.
• 203 smokers who slipped in the first 6 weeks of quitting recorded their feelings on handheld computers at each slip. 79% of slips were one cigarette or less.
• A separate study of 1,190 smokers found that those who slipped in the first 8 weeks were less often off cigarettes at 6 months, 21.6% against 70.7%.
• In a mock app modelled on Duolingo, an intact streak was four answers in a row and a broken one was three, then a message that the fourth question was full. Half saw their run marked in an in-app log, half answered in a plain survey.
• Students took a 10 item vocabulary test hard enough that they averaged 4 out of 10. Before a second test they could study for as long as they liked, and a short statement added to the instructions changed by group.

What it cannot tell you

• The quitting studies are about smoking, and the streak and kindness studies used mock app tasks and students. None of it is about pornography.
• Quit attempts were self-reported, with no test to confirm who had quit. The 29.6 is an average across a population and cannot predict how many tries any one person needs.
• Six weeks of tracking, and a strict definition of relapse. The authors cite older studies in which 85% to 95% of smokers who slip eventually relapse. The smokers were heavy smokers who sought treatment, mostly white and better educated than average.
• The streak experiments used a few questions, games or exercises in mock apps, with streaks of 3 or 4, run with paid online volunteers between 2017 and 2021. None of it involved days of recovery.
• The kindness study used groups of 27 to 30 students at one university. Only the gap with no message was significant, and retest scores did not differ significantly by group.

In P-Minus

P-Minus has no streak or day counter. It tracks constancy, and a lapse never resets anything. A lapse opens a repentance flow built on mercy and adds one repentance ritual to the day, with no penance. You renew your intention, the flow ends by asking what you will do at that cue next time and saves your answer as a plan, and the next day runs as normal.

Sources

1. Chaiton M, Diemert L, Cohen JE, et al. Estimating the number of quit attempts it takes to quit smoking successfully in a longitudinal cohort of smokers. BMJ Open. 2016;6(6):e011045.
2. Kirchner TR, Shiffman S, Wileyto EP. Relapse dynamics during smoking cessation: recurrent abstinence violation effects and lapse-relapse progression. J Abnorm Psychol. 2012;121(1):187-97.
3. Deiches JF, Baker TB, Lanza S, Piper ME. Early lapses in a cessation attempt: lapse contexts, cessation success, and predictors of early lapse. Nicotine Tob Res. 2013;15(11):1883-91.
4. Silverman J, Barasch A. On or off track: how (broken) streaks affect consumer decisions. J Consum Res. 2023;49(6):1095-117.
5. Breines JG, Chen S. Self-compassion increases self-improvement motivation. Pers Soc Psychol Bull. 2012;38(9):1133-43.

Sources checked October 2026.