Most period-tracking apps predict ovulation from a calendar — and get it wrong often

If a period-tracking app tells you when you ovulate by counting days on a calendar — which most of them do — it is frequently wrong. In an analysis of 543,167 cycles, calendar-based apps misclassified at least one genuinely fertile day as "safe" in 67% of cycles, and the problem was worse in women with irregular cycles [s1]. The apps that do better are the ones that measure something biological rather than just counting.

Why calendar prediction fails

The core issue is that ovulation does not keep to a schedule. The popular mental model — ovulation on day 14 of a 28-day cycle, a predictable fertile window around it — is an average, not a rule, and the day of ovulation varies from cycle to cycle even in regular menstruators. An app that predicts the fertile window purely from past cycle dates is extrapolating from that average, so when a real cycle deviates, the prediction misses.

The 2026 analysis quantified the consequence by comparing app predictions against fertile days confirmed with urinary luteinising-hormone tests — an actual biological marker of impending ovulation [s1]. Calendar-based tracking left 67% of cycles with at least one fertile day wrongly flagged as safe, translating to a modelled pregnancy risk of 22% (rising to 65% for irregular cycles) [s1]. Temperature-supported tracking did better but was still substantial, with 41% of cycles containing a misclassified fertile day and a modelled risk of 9% [s1]. A regulated, FDA-cleared digital contraceptive — which uses temperature readings within a validated algorithm — left only 3% of cycles with any at-risk day, and a modelled risk of 0.5% [s1]. The gap between a calendar app and a regulated device is the gap between guessing and measuring.

This is not new, and it is not one bad app

The finding has a decade of consistency behind it. In 2016, researchers standardised a textbook 28-day cycle and checked how well fertility web sites and apps predicted its fertile window [s2]. Of 20 web sites and 33 apps, only one web site and three apps predicted the precise fertile window; all of them included the single most fertile day, but the range they gave for the surrounding window varied widely [s2]. The authors concluded that these tools are "generally inaccurate," while noting that the clinical effect of that inaccuracy was unknown at the time [s2].

The reason the problem is so widespread is the composition of the app market. A 2020 audit reviewed 90 fertility-tracking apps and found that most — 54.4%, or 49 apps — tracked nothing but menstrual cycle dates, with the remaining 45.6% incorporating at least one fertility-awareness measure such as basal body temperature, cervical mucus or luteinising hormone [s3]. The audit's summary was that "unreliable calendar apps remain the most commonly available fertility apps on the market," in a category it described as largely unregulated [s3]. In other words, the least accurate method is the default a typical user encounters.

What this does and does not mean

For someone trying to conceive, an inaccurate fertile-window prediction mostly wastes effort and time, and the 2016 authors were careful to say the clinical impact was unquantified [s2]. For someone relying on an app to avoid pregnancy, the stakes are different, and the 2026 analysis is explicit: calendar-based and temperature-supported period-tracking apps "should not be considered reliable tools for pregnancy prevention," which is a distinct product category from the regulated digital contraceptives it outperformed them against [s1]. The site covers the effectiveness of app-based contraception separately.

Two limits keep this honest. The 2026 analysis modelled pregnancy risk from misclassification rather than counting actual pregnancies, and it drew on data from a fertility-app company [s1]. The 2016 study tested a single idealised cycle, not real-world use [s2]. But across a modelled half-million cycles, a standardised audit and a market survey, the direction is the same: a calendar-based app is a convenient record of your periods, not an accurate predictor of when you ovulate.

This article is informational and is not medical advice. Anyone relying on cycle tracking to achieve or avoid pregnancy should discuss the method's reliability with a clinician.