Q&A: How do I calculate ovulation with long cycles?

The variation responsible for cycle length is the proliferative phase of the cycle--that half from the period to ovulation. For the most part, the time from ovulation to the next period is consistently around 14 days, assuming it's an adequate ovulation.* Unfortunately, this makes it hard to predict ovulation since only a past ovulation can be calculated accurately "in arrears." Yes, the second half of the cycle, the "secretory phase," is consistently two weeks, and this is a favorite question on the written board exams in OBGYN.

*It's of academic interest that there's such a thing called an "inadequate corpus luteal phase," in which there's not enough progesterone put out from the corpus luteum (that bed of tissue from which ovulation took place). But I think this reflects on the poor quality of the egg or ovulation that occurred, not the corpus luteum; therefore, if it can be documented with blood tests that the second half of the cycle is longer than 14 days, this probably represents a faulty or even a non-ovulation.