Someone posts a study. Someone else replies that it does not apply. Both are usually right, because they are not discussing the same practice. NoFap, semen retention, hard mode and celibacy name four different commitments, and the studies quoted in the argument are usually measuring none of them exactly.
The four words, as practitioners use them
- NoFap. Abstaining from pornography and masturbation, with the original emphasis on porn.
- Semen retention. Avoiding ejaculation by any route, partnered sex included, and typically carrying a claim that seminal fluid itself holds something worth conserving.
- Hard mode. No porn, no masturbation and no orgasm of any kind, including with a partner.
- Celibacy. In forum usage, abstaining from partnered sex. Used that way it says nothing at all about masturbation.
The edges are where the difference shows. A man on hard mode who shares a bed with his partner is not celibate in the ordinary sense. In the forum sense of the word, a celibate man may masturbate daily. Retention and hard mode part company over one detail: orgasm without ejaculation. That phenomenon is documented, thinly. An interview study of 21 men who described themselves as multiply orgasmic reported that detumescence does not always follow orgasm, that a non-ejaculatory orgasm can occur before as well as after an ejaculatory one, and that a series of orgasms is possible. It is small, self-selected, descriptive and from 1989, so it establishes that the phenomenon occurs in some men and nothing about how common or how learnable it is. It is still enough to show that "no orgasm" and "no ejaculation" are two different rules.
Celibacy is the one word in that list that is not native to these forums, and it is worth keeping separate. Inside the religious traditions that ask for it, celibacy is a vow taken as part of a whole way of life, with its own reasoning, its own obligations and its own vocabulary. It is not a rule about a fluid and it is not a counter. Borrowing the word as a streak category does not make the two things the same, and this article uses it only in the narrow forum sense above.
NoFap, semen retention, hard mode and celibacy are basically the same thing.
The four words name four different commitments, and the communities using them distinguish them, inconsistently. The distinction matters most for reading evidence: the published literature is overwhelmingly about pornography abstinence. The Fernandez 2021 thematic analysis of 104 abstinence journals and the Fernandez 2023 randomised trial (n = 176) both study porn abstinence. The Zimmer and Imhoff survey (n = 1,063) studies masturbation abstinence. Almost nothing studies retention as retention practitioners define it, so a result about one practice is usually not evidence about another in either direction.
Two wings, two vocabularies
The vocabulary is unsettled because the movement split. One wing is secular and anti-porn. Its frame is compulsion and recovery, its language is rebooting, relapse and withdrawal, and its target is the screen rather than the fluid. A thematic analysis of 104 public abstinence journals written by male forum members found that people began rebooting to address perceived addiction and perceived sexual difficulties, and organised their accounts around four themes: abstinence is the solution, abstinence sometimes seems impossible, abstinence is achievable with the right resources, and abstinence is rewarding if you persist.
The other wing is spiritual and concerns the semen itself. A 2025 analysis of NoFap lifestyle gurus on YouTube describes a construction in which masculine energy is a life force partially contained in semen: precious, scarce, and capable of elevating men into a dominant minority if preserved. The same paper documents how the practice is monetised, with micro-celebrity accounts run as businesses. That is a description of how the belief travels and is sold. It is not a claim about physiology in either direction.
Where the modern version comes from, and what this article does not assess
Within the modern anglophone version, chapter 11 of Napoleon Hill's Think and Grow Rich, published in 1937 and titled The Mystery of Sex Transmutation, is the text most often invoked: it argues that sexual desire is a creative force that can be redirected toward achievement. It contains no data, no measurement and no control condition. Calling it the origin of the practice would be wrong. Sexual continence has far older and quite separate histories: brahmacharya in Hindu and Jain thought, the conservation of jing in Daoism, shukra and ojas in Ayurveda, and the purity movements of nineteenth-century Britain and America. Each of those is a system with its own reasoning, its own vocabulary and its own obligations, each deserves to be read on those terms rather than as early self-improvement content, and none of them is assessed here. This piece is about the modern anglophone claims and the modern anglophone evidence, and that limit is worth stating rather than hiding. The movement's own founding story, its dates and its personalities, is told mostly by the movement about itself, and nothing here rests on it.
The argument is rarely about the evidence. It is about which practice the evidence was measuring.
Different practices change different variables
Porn abstinence changes what you look at. Masturbation abstinence changes what you do alone. Retention changes what leaves the body, including inside partnered sex. Celibacy changes your relationships. These are not degrees of one intervention, so a result about one is usually not evidence about another.
Withdrawal is the clearest case. There is one randomised controlled test of short-term pornography abstinence: 176 undergraduates in Malaysia, 64.2 percent female, mean age 21.4, randomised to seven days of abstinence (n = 86) or to carrying on as usual (n = 90), with daily surveys. The preregistered prediction that abstainers would show withdrawal-related symptoms was not supported, with no significant group effects on craving, positive affect, negative affect or withdrawal symptoms. Compliance was self-reported, and 45.35 percent of the abstinence group lapsed at least once inside the week.
Quitting porn produces withdrawal symptoms, which proves it is an addiction.
The single randomised test found no group-level withdrawal over seven days, and that is all it can carry: n = 176, majority female, undergraduate, non-clinical, self-reported compliance, nearly half the abstinence arm lapsing, and pornography rather than ejaculation. It is not a demonstration that men abstaining for weeks experience nothing, and the study's own exploratory analysis suggested craving effects might appear where problematic use was high and use was daily. The one controlled test went against the prediction, and the question remains open.
The retention wing's central number has a different problem. The claim that testosterone peaks on day seven at 145.7 percent of baseline comes from a 2003 paper with 28 volunteers in the Journal of Zhejiang University Science A, retracted in 2021 because it substantially overlapped an earlier Chinese-language article by the same author. Two of its authors agreed to the retraction. A retracted paper is not evidence that the day-seven peak is real, and it is not evidence that it is false. It was never replicated, and almost nobody repeating the figure knows it was withdrawn.
Where retention has an outcome literature of its own
Because retention is defined by ejaculation rather than by screens, it inherits a body of evidence that porn abstinence does not, and that evidence does not run the way the depletion story predicts. In the Health Professionals Follow-up Study, 31,925 men were followed across 480,831 person-years with 3,839 prostate cancer diagnoses. Ejaculating 21 or more times per month at ages 20-29 and at ages 40-49 was associated with 19 percent and 22 percent lower risk of a prostate cancer diagnosis respectively, compared with 4-7 times per month. On fertility, a review of the abstinence-interval literature concludes that shorter intervals are associated with better sperm function and recommends ejaculating every one to two days for men attempting conception.
Ejaculating drains you and depletes the body of something you need to conserve.
No study has demonstrated depletion of any measured resource at ordinary ejaculation frequencies, and on the one hard health outcome studied at scale the association runs opposite to what the depletion model predicts: n = 31,925, 480,831 person-years, 3,839 diagnoses, with 21 or more ejaculations a month at ages 20-29 and 40-49 associated with 19 percent and 22 percent lower risk of a prostate cancer diagnosis than 4-7 times a month. That cohort is observational, frequency was self-reported and partly retrospective, the participants are health professionals rather than a general population, and detection bias cannot be excluded. It does not show that ejaculating prevents anything, and it says nothing about what will happen to any individual. It is also the largest relevant dataset that exists, and it does not point where the depletion claim predicts.
The compulsive sexual behaviour question, fairly
A diagnosis exists, and it says close to the opposite of what it is usually quoted as saying. Compulsive Sexual Behaviour Disorder entered the ICD-11 as code 6C72: a persistent failure to control intense repetitive sexual impulses, sustained over six months or more, with marked distress or impairment. The working group placed it among impulse control disorders and explicitly not among disorders due to substance use and addictive behaviours, on the stated grounds that there is not yet definitive information on whether the underlying processes are equivalent. The criteria also state that distress arising entirely from moral judgement of one's own sexual impulses is not sufficient for the diagnosis. The DSM-5 and DSM-5-TR contain no equivalent. The underlying complaint is not rare: in a nationally representative US survey of 2,325 adults, 8.6 percent of adults and 10.3 percent of men reported clinically relevant distress or impairment around difficulty controlling sexual feelings, urges and behaviours. That survey measured distress, not addiction, and says nothing about causes.
The WHO officially recognised porn addiction, so the science is settled.
A real classification exists: compulsive sexual behaviour disorder, ICD-11 code 6C72. The WHO placed it among impulse control disorders and specifically declined to classify it as an addiction, citing insufficient evidence that the mechanisms match substance use, gambling or gaming. The criteria also exclude distress arising purely from moral disapproval of one's own behaviour, and the DSM-5 and DSM-5-TR contain no equivalent diagnosis. The second half of the claim is false on the working group's own reasoning: unsettledness is the stated reason the disorder was filed where it was.
Two camps, both still publishing
Researchers genuinely disagree, and both camps publish in serious journals. Brand and colleagues, authors who broadly favour the addiction framing, argue that problematic pornography use has some support from self-report, clinical and behavioural evidence for the ICD-11 residual category covering other specified disorders due to addictive behaviours, while noting the evidence base is uneven. Ley, Prause and Finn, the most prominent public critics of that model, argue in an explicitly polemical narrative review that the addiction framing rests on poor experimental design, limited methodological rigour and lack of model specification, and that the phenomena are better accounted for by high sex drive, sensation seeking, desire discrepancy within couples, and conflict driven by religiosity. Neither paper closed the question, both groups are still working, and a position attributed to any of these researchers should be checked against their current output rather than a decade-old citation.
One finding here is unusually robust, and it is the one least often quoted in practitioner spaces. A systematic review and meta-analysis found that moral incongruence about pornography use, meaning the experience of behaving against one's own beliefs, is associated with greater distress about use, greater general psychological distress, more reported problems and higher perceived addiction, independent of how much a person actually uses. Across three further samples including a four-wave one-year longitudinal one, self-reported problematic use tracked erectile difficulty while use volume itself largely did not, with no evidence of causal links in the longitudinal data.
If a man says his use is out of control, that tells you his use is heavy.
The strongest predictor of feeling addicted is disapproving of the behaviour, not the amount of it. The constituent studies are overwhelmingly cross-sectional, self-report, and weighted toward US and religious samples where moral incongruence is most measurable. This is not a claim that the distress is fake or that impaired control does not exist: it is a claim about what these self-report scales are partly measuring, which is a person's judgement of himself.
What this means if you are the one doing it
If you cannot say which of the four commitments you are making, you cannot say which claims are even addressed to you. The definitional questions have no external answer. Whether a nocturnal emission breaks a streak depends entirely on whether your practice is about a substance or about volition, and both positions are internally consistent. No study is going to settle that, because it is a question about your own aim.
Research can tell you whether semen retention works.
No randomised or controlled trial of semen retention as a practice, on any psychological outcome, could be located in the peer-reviewed literature. The nearest prospective data comes from a study of No Nut November participation: 435 respondents before the month and 114 after, roughly three quarters lost between timepoints, self-selected and self-reported throughout. Over the abstinence period no measures of sexual wellbeing significantly differed between participants and non-participants, and the authors conclude that a period of abstinence may not be associated with either improved or worsened sexual wellbeing.
There is also a structural reason a population study may never resolve this. Men who start retention almost never start only retention: the same week typically brings training, an earlier bedtime, less alcohol, fewer apps, and a new community that expects improvement. Each of those has documented effects on mood, sleep and self-rated confidence, and no study has attempted to separate them. Randomising is the usual fix, but nobody can be blinded to whether they are ejaculating, so expectancy stays in the design permanently. Tracking is not neutral either: a meta-analysis of 138 randomised studies with 19,951 participants found that prompting people to monitor progress toward a goal improves performance and goal attainment, with larger effects when progress is recorded. Tracking probably helps, which is exactly why tracking is also a confound. Within one person, changing one variable at a time and measuring repeatedly is a recognised methodology, set out formally in AHRQ's guide to n-of-1 trials. It cannot say what a practice does in general. It can say what changed for one person, when, and alongside what else.
You cannot blind a man to whether he is ejaculating.
One caution belongs in the body of this piece rather than in a footnote. The symptom picture people describe during these practices, flat mood, low libido, loss of motivation, fatigue, overlaps with the picture of depression, and community lore does not distinguish between them. Persistent low mood, anhedonia, pain, or erectile problems that do not resolve are worth taking to a doctor rather than reading as a stage in a protocol.
Say which experiment you are running
Before starting, write the commitment down in plain terms: what you are abstaining from, what counts as a break, and what you expect to change. That one paragraph does more for the quality of your own conclusions than any study currently available, because it determines whether the evidence you read is about you at all.