A textbook menstrual cycle lasts 28 days, but real-world data from hundreds of thousands of cycles shows the average is closer to 29 days, with most people falling somewhere between 21 and 35 days. More interesting than the total number, though, is how unevenly that time is divided among the cycle’s phases. The follicular phase and luteal phase split the cycle into two halves of surprisingly different character: one is flexible and prone to shifting, while the other is relatively fixed. Understanding where each phase begins and ends, and how much each one can stretch or shrink, tells you far more than a single cycle-length number ever could.
The Follicular Phase
The follicular phase starts on the first day of your period and runs until ovulation. During this stretch, your brain’s pituitary gland releases follicle-stimulating hormone, which prompts several follicles in the ovaries to begin developing. Eventually one follicle becomes dominant and produces rising levels of estrogen, which thickens the uterine lining and ultimately triggers the hormonal surge that causes ovulation.
In a large analysis of more than 600,000 ovulatory cycles tracked through a fertility app, the mean follicular phase length was 16.9 days, but with enormous spread: the middle 95% of values ranged from 10 to 30 days.1npj Digital Medicine. Real-world menstrual cycle characteristics of more than 600,000 menstrual cycles That range is the single biggest reason menstrual cycles vary so much in total length. When a cycle runs long or short, it is almost always the follicular phase doing the stretching or compressing. A cycle of 24 days and a cycle of 34 days can both have perfectly healthy luteal phases; the difference sits almost entirely in how long it took for a dominant follicle to mature.
Several lifestyle factors nudge follicular phase length in one direction or another. In a study of 201 women, recent use of oral contraceptives and duration of prior pill use were both correlated with a longer follicular phase. Occasional marijuana use (up to three times in the previous three months) was linked to a follicular phase about 3.5 days longer than in nonusers, while a history of miscarriage was associated with a follicular phase roughly 2 days shorter.2PubMed Central. Lifestyle and reproductive factors associated with follicular phase length These shifts may not sound dramatic, but in a phase that averages 17 days, a few days in either direction represents a meaningful change.
Ovulation
Ovulation itself is the shortest event in the cycle, lasting roughly 12 to 24 hours. It happens when a surge of luteinizing hormone causes the dominant follicle to rupture and release an egg into the fallopian tube. Because the egg survives only about a day after release, the actual fertile window around ovulation depends more on how long sperm can survive (up to five days in the reproductive tract) than on the ovulation event itself.
There is no fixed “day 14” for ovulation. Because the follicular phase varies so widely, ovulation can happen anywhere from around day 10 to day 30 or later. This is one of the most commonly misunderstood points about the menstrual cycle, and it matters enormously for anyone relying on calendar-based methods to either achieve or avoid pregnancy.
The Luteal Phase
Once the egg is released, the ruptured follicle transforms into a temporary structure called the corpus luteum, which pumps out progesterone. Progesterone stabilizes the uterine lining and prepares it for a possible embryo. If no pregnancy occurs, the corpus luteum breaks down after roughly 12 to 14 days, progesterone drops sharply, and the lining sheds as your period.3PubMed Central. The inadequate corpus luteum
The large cycle-tracking study of 600,000-plus cycles found a mean luteal phase of 12.4 days, with 95% of values falling between 7 and 17 days.1npj Digital Medicine. Real-world menstrual cycle characteristics of more than 600,000 menstrual cycles Compare that 10-day spread with the follicular phase’s 20-day spread: the luteal phase is strikingly more consistent. This relative stability is the basis for certain “count back from your next period” fertility methods. If you know when your period starts, you can estimate that ovulation occurred roughly 12 to 14 days earlier. The method isn’t perfect, but it works better than guessing forward from the start of your last period, because the follicular phase is the unpredictable half.
Menstrual Bleeding
Menstruation, the bleeding phase, overlaps with the early follicular phase. It begins when progesterone withdrawal causes the uterine lining to break down and shed. The duration of bleeding varies, but studies consistently place the average around 4 to 5 days.4PubMed Central. Influence of Socioeconomic Status in the Age at the of Menarche and Duration of Menstrual Bleeding Bleeding for 2 to 7 days is generally considered within the normal range. The heaviest flow typically occurs on days 1 and 2, then tapers off. Periods that regularly last longer than 7 days or involve very heavy bleeding may signal conditions worth investigating with a healthcare provider, though a single unusually long or heavy period is common and not necessarily a sign of a problem.
Why Your Cycle Probably Isn’t 28 Days
The 28-day cycle is a useful teaching model, but it is not particularly representative. In a retrospective study of nearly 1.6 million women using a cycle-tracking app, only about 16% had a median cycle length of exactly 28 days. Roughly the same proportion had a 27-day or 29-day median.5PubMed Central. Menstrual Cycle Length and Patterns in a Global Cohort of Women Using a Mobile Phone App: Retrospective Cohort Study A separate study found a mean cycle length of 27.7 days with a 95% range spanning 23 to 32 days.6Fertility and Sterility. The normal variabilities of the menstrual cycle
Perhaps more surprising than cycle-to-cycle variation between people is the variation within a single person. Data from the Apple Women’s Health Study found that the average within-individual variability was 4 to 6 days across most age, ethnicity, and body-mass categories.7PubMed Central. Menstrual cycle length variation by demographic characteristics from the Apple Women’s Health Study That means even someone with a 29-day average might routinely see cycles ranging from 26 to 33 days. Having cycles that bounce around within that window is normal, not a sign that something is wrong.
How Age Reshapes the Cycle
Cycle length changes meaningfully across a person’s reproductive years. In the early years after menarche, cycles tend to be longer and more erratic because the hormonal feedback system is still maturing. By the mid-to-late 30s, cycles tend to become their most regular and often shorten slightly, mainly because the follicular phase compresses as ovarian reserve declines. The Apple Women’s Health Study found that the 35-to-39 age group had the lowest cycle-length variability, with relative increases in variability seen in both younger and older groups.7PubMed Central. Menstrual cycle length variation by demographic characteristics from the Apple Women’s Health Study
As people approach perimenopause (typically in the mid-40s), cycle lengths often become erratic again. Cycles may lengthen, shorten, or alternate between the two before eventually stopping at menopause. The total span of reproductive years is surprisingly influenced by when menarche began: in a large Norwegian population study, for each one-year increase in age at menarche, the median reproductive period shortened by about a year.8PubMed Central. The relation of age at menarche with age at natural menopause: a population study of 336 788 women in Norway
Stress, Energy Deficits, and Shift Work
Chronic psychological stress can disrupt cycle timing by interfering with the brain’s signaling to the ovaries. The pathway runs through a group of neurons that control the pulsing release of gonadotropin-releasing hormone, which in turn governs the hormonal cascade behind ovulation. When the stress response is activated persistently, those pulses can slow or become irregular, delaying ovulation or preventing it entirely.9PubMed Central. Chronic Stress and Ovulatory Dysfunction: Implications in Times of COVID-19 Because the follicular phase is the flexible portion of the cycle, stress-related delays typically show up as a longer follicular phase, pushing the whole cycle later.
Energy availability matters too. When calorie intake consistently falls short of what the body burns, especially in combination with intense exercise, the hypothalamus can dial down reproductive hormones. A study randomizing exercising women to varying energy deficits found that the magnitude of daily energy shortfall predicted how frequently menstrual disturbances occurred.10American Journal of Physiology-Endocrinology and Metabolism. Magnitude of daily energy deficit predicts frequency but not severity of menstrual disturbances associated with exercise and caloric restriction In milder cases this means a longer cycle; in severe cases it can mean periods stop altogether.
Circadian disruption from rotating shift work is another well-documented influence. In a study of nurses, women who performed more than seven night shifts per month had shorter cycles, and the proportion of nurses with a cycle between 25 and 31 days dropped from about 82% to 68% after switching to shift work. Over a two-year follow-up, cycle length did not fully recover.11PubMed Central. Rotating shift work and menstrual characteristics in a cohort of Chinese nurses A larger study found a dose-response relationship: the more months of rotating shift work, the greater the likelihood of very short cycles (under 21 days) or very long ones (40-plus days).12PubMed Central. Rotating shift work and menstrual cycle characteristics The effect appears to work through disruption of melatonin and other clock-driven signals that help coordinate reproductive hormones.13PubMed Central. Shift work and circadian dysregulation of reproduction
When Ovulation Doesn’t Happen
Not every cycle includes ovulation, even in people with seemingly normal periods. In a population-based study from Norway, researchers measured progesterone during the expected luteal window of spontaneous, normal-length cycles. About 37% of those cycles turned out to be anovulatory, meaning no egg was released despite the person bleeding on a regular schedule.14PubMed Central. Ovulation Prevalence in Women with Spontaneous Normal-Length Menstrual Cycles – A Population-Based Cohort from HUNT3, Norway That number is higher than many people expect, in part because anovulatory cycles can still produce withdrawal bleeding that looks and feels like a normal period.
A year-long prospective study of healthy women who had been prescreened for normal cycle and luteal-phase lengths found that 17% experienced at least one anovulatory cycle during the year, and 55% had at least one cycle with a short luteal phase (under 10 days).15Human Reproduction. Prospective 1-year assessment of within-woman variability of follicular and luteal phase lengths in healthy women prescreened to have normal menstrual cycle and luteal phase lengths In other words, occasional deviations from a “textbook” ovulatory cycle are common even among people with regular periods. The practical takeaway is that if you are tracking cycles for fertility purposes, confirming ovulation directly (through temperature shifts, hormone tests, or both) is far more informative than just counting cycle days.
Short Luteal Phases and Fertility
A luteal phase that consistently runs short raises questions about fertility, because the uterine lining may not have enough time under progesterone’s influence to support implantation. In a prospective study of women trying to conceive, a short luteal phase (defined as 11 days or less including ovulation day) occurred in about 18% of observed cycles. Women with a short luteal phase had somewhat lower odds of pregnancy in the cycle immediately following, though the difference was not statistically significant. After 12 months of trying, cumulative pregnancy rates were similar regardless of whether early cycles had been short.16PubMed Central. A prospective evaluation of luteal phase length and natural fertility
This is somewhat reassuring: a single short luteal phase, or even a few of them, does not necessarily mean you will have trouble getting pregnant. Persistent short luteal phases over many consecutive cycles are a different story and may warrant investigation, but the occasional short one appears to be part of normal variation. It is worth noting that definitions of “short” vary across studies, with some using 10 days and others 11, which makes comparing research findings tricky.
How Contraception Reshapes the Cycle
Hormonal contraception fundamentally alters the natural cycle by suppressing ovulation and replacing the body’s own hormonal fluctuations with steady exogenous hormones. Combined oral contraceptives create an artificial 28-day cycle with a hormone-free interval that triggers withdrawal bleeding. This bleeding is pharmacologically scheduled, not the result of a corpus luteum breaking down, so the concept of a follicular or luteal “phase” doesn’t apply in the usual sense.
Breakthrough bleeding, meaning unscheduled spotting or bleeding while taking active hormones, is one of the most common reasons people discontinue contraception. With combined pills, breakthrough bleeding happens because of hormonal fluctuations or insufficient stabilization of the uterine lining. With progestin-only methods, it tends to result from the continuous progestin exposure itself changing the endometrium’s blood-vessel structure.17PubMed Central. Understanding Problematic Bleeding When Using Contraception: Guidance for Clinicians After stopping hormonal contraception, it can take several cycles for natural phase lengths to normalize. Some people resume regular ovulatory cycles within a month; others take three to six months or occasionally longer.
Tracking Your Phases at Home
Several methods exist for estimating where you are in your cycle, each with trade-offs in accuracy and convenience. Basal body temperature (BBT), taken every morning before getting out of bed, typically rises by about 0.2 to 0.5°C after ovulation and stays elevated through the luteal phase. The problem is that BBT confirms ovulation only after it has already happened, making it more useful for understanding past cycles than predicting the current fertile window in real time.
Urine-based luteinizing hormone (LH) test kits detect the hormone surge that precedes ovulation by about 24 to 36 hours. One study found that LH kits correctly identified the ovulatory window in 93% of cycles, compared to 72% for BBT and 61% for cervical mucus observation.18PubMed. Simple office methods to predict ovulation: the clinical usefulness of a new urine luteinizing hormone kit compared to basal body temperature, cervical mucus and ultrasound Cervical mucus monitoring, while less precise on its own, adds useful complementary information when combined with other methods.
More recently, wearable devices that continuously track temperature and heart rate have been paired with machine-learning algorithms to predict the fertile window. In one study, such an algorithm achieved about 87% accuracy in predicting the fertile window for regular menstruators, though accuracy dropped to roughly 73% for those with irregular cycles.19PubMed Central. Tracking of menstrual cycles and prediction of the fertile window via measurements of basal body temperature and heart rate as well as machine-learning algorithms Newer approaches using electrical impedance measurements of cervical mucus have shown improved sensitivity and specificity compared to BBT alone, though these remain primarily research tools for now.20PubMed Central. The effectiveness of cervical mucus electrical impedance compared to basal body temperature to determine fertility window
Environmental Chemicals and Phase Timing
Certain environmental pollutants can shift phase lengths in ways that may fly under the radar. Organochlorine pesticides, including DDT and its breakdown product DDE, have been associated with decreased progesterone levels and a shorter luteal phase. Women in the highest exposure quartile for DDT had luteal phases roughly 1.5 days shorter than those with lower exposure.21PubMed Central. Exposure to endocrine disruptors during adulthood: Consequences for female fertility While DDT has been banned in many countries for decades, it persists in the environment and remains in use in some regions for malaria control. Other endocrine-disrupting chemicals, including certain plasticizers and industrial compounds, are under active investigation for similar effects, though the evidence for most individual chemicals is still emerging.
Thyroid function is another factor sometimes blamed for cycle irregularities, but the relationship appears weaker than widely assumed. A longitudinal study of premenopausal women found that thyroid hormone levels were not meaningfully associated with cycle or phase lengths. While there was a subtle trend toward longer cycles with higher free T4, the differences were small and the confidence intervals overlapped substantially, with the variation driven by the follicular phase rather than the luteal phase.22PubMed Central. Thyroid hormones and menstrual cycle function in a longitudinal cohort of premenopausal women Clinically significant thyroid disease can certainly disrupt cycles, but mild, subclinical variation in thyroid levels does not appear to be the common culprit that popular health advice sometimes makes it out to be.