Dating Theory Calculator (37% Rule)
This calculator applies the optimal stopping theory (the famous "37% rule") to dating. Enter how many first dates you are willing to go on and how demanding you want to be, and the calculator tells you exactly how many to treat as a learning phase before you seriously consider settling down. It also shows your probability of finding your ideal match, the risk that your perfect person appeared too early, and how rejection affects your odds. The math comes from the secretary problem, one of the most celebrated results in probability theory.
What is the dating theory (optimal stopping) calculator?
The dating theory calculator applies the secretary problem, one of the most celebrated puzzles in probability theory, to the search for a long-term partner. The core question is: if you are going to meet a fixed number of potential partners one by one and cannot go back to someone you rejected, when should you stop exploring and commit? The answer is the 37% rule: date the first N/e people (roughly 37% of your total lifetime dating pool) without committing to any of them, using that time purely to calibrate your expectations. After that exploration phase, commit to the first person who is better than everyone you have seen before. Mathematical analysis, first published by Merrill in the 1950s and later formalised by Gilbert and Mosteller in 1966, shows that this strategy gives you roughly a 36.8% (1/e) chance of finding the single best match, which is the highest probability achievable by any strategy.
How the 37% rule works in practice
Suppose you expect to go on 20 first dates over your dating life. The calculator tells you to treat the first 7 or 8 as a learning window: date them, enjoy them, but do not commit. Use that sample to build a mental benchmark of what a great partner looks like for you personally. From date 8 or 9 onward, you switch to selection mode: the moment someone comes along who is better than everyone in your learning window, you commit to them. If no one surpasses your benchmark, you end up with whoever you meet last. This rule is surprising because 37% feels like you are wasting time, but the math shows it is the optimal trade-off between gathering information and acting on it. Going too short means you commit before you know what good looks like; going too long means you have already passed over your best option.
Adjusting for real-world complications
The classic secretary problem has several simplifying assumptions that do not hold in real dating. First, you might not be aiming for the single best person, you might be happy with any of your top three or top five, which substantially raises your success probability for the same strategy. Second, the person you choose might not choose you back, so the rejection risk slider lets you see how mutual-choice affects your odds. Third, your dating pool is not fixed: people come and go, your own preferences change, and some people are simply unavailable when you are ready to commit. The calculator gives you a mathematically grounded starting point, not a rigid life script. Think of the exploration-phase number as an evidence-based minimum, not an exact count.
Limitations and what the math does not capture
Optimal stopping theory treats each candidate as independently ranked by a single preference ordering. Real attraction is multidimensional, shifts over time, and depends on context. The model also assumes you know your pool size in advance and that you can recall past candidates perfectly. In practice, dating pools are open-ended, people from your past do sometimes reconnect, and your preferences evolve. Research in psychology (Todd and Miller, 1999) shows that people tend to set their aspiration level from early experiences and use a "satisficing" heuristic rather than strict optimal stopping. The 37% rule is best understood as a lower bound on how much exploration you should do before committing, not as a precise prescription.
Optimal cutoff and success probability by pool size
| Total dates (N) | Explore-phase cutoff (M) | Cutoff as % of N | Success probability |
|---|---|---|---|
| 5 | 2 | 40% | 43% |
| 10 | 4 | 40% | 40% |
| 15 | 6 | 40% | 39% |
| 20 | 7 | 35% | 38% |
| 30 | 11 | 37% | 37% |
| 50 | 18 | 36% | 37% |
| 100 | 37 | 37% | 37% |
| 200 | 73 | 37% | 37% |
| 500 | 184 | 37% | 37% |
For the single-best-match goal, the optimal strategy always rejects about 37% of candidates regardless of pool size. Success probability converges to about 36.8% (1/e).
Frequently asked questions
What is the 37% rule in dating?
The 37% rule says you should date the first 37% of your total expected lifetime partners without committing to any of them, then commit to the first person thereafter who is better than everyone in that initial sample. The "37%" comes from 1/e (where e is Euler's number, approximately 2.718). Following this rule gives you about a 36.8% chance of finding your single best match, which is the highest probability any pure strategy can achieve.
What is the optimal stopping problem in the context of dating?
The optimal stopping problem (also called the secretary problem) asks: given a sequence of candidates observed one at a time, where you must accept or reject each immediately and permanently, when should you stop and choose? In dating, each candidate is a potential partner. The goal is to maximise the probability of choosing the best one. The mathematically optimal answer is to reject the first N/e candidates and then pick the next one who beats all of them.
Does the 37% rule actually work in real life?
It works as a theoretical benchmark, but real dating violates most of the model's assumptions. You cannot know your true pool size, rejection is mutual (your chosen partner might not want you), people from the past can re-enter your life, and preferences change over time. Studies suggest people naturally use shorter exploration windows than the model prescribes. The rule is most useful as a reminder that some intentional learning phase before committing is mathematically justified, and that both committing too early and holding out indefinitely for perfection reduce your odds.
How does accepting a top-3 or top-5 match change my odds?
If you are willing to be happy with any of your top three matches rather than strictly the single best, your success probability rises substantially. For large pools, accepting a top-k match gives a theoretical success probability approaching 1 - (1/e)^k. For k = 3, that is about 1 - 0.050 = 95%, compared to about 37% for the single best only. In practice the gains are smaller because the model is approximate, but broadening your criteria even slightly has a large effect on your odds.
What should I do if I am already past the 37% threshold?
If you have already dated more people than the optimal cutoff, you are in selection mode right now. The optimal strategy for you is to commit to the next person you date who is better than everyone you have dated so far. The earlier exploration was not wasted: it gave you a benchmark. The key insight is that you should not restart the clock or wait longer, commit to the next person who genuinely exceeds your past experience.
Does the rejection risk from my chosen partner affect the strategy?
Yes. If there is a chance your chosen partner will not want to commit to you, your effective success probability drops. The pure optimal stopping model ignores this, but the calculator accounts for it with a rejection probability slider. At 0% rejection risk, your success probability is about 37%. At 25% rejection risk, it falls to about 28%. The optimal cutoff point (how many to reject before switching to selection mode) does not change much, but your realistic odds of a successful outcome decrease with higher rejection risk.
Is the dating theory calculator the same as a compatibility calculator?
No, they are very different. A compatibility calculator tries to predict how well two specific people will get on based on personality traits, values, or interests. The dating theory calculator does not assess any individual pair at all; it tells you how many people to date before you are ready to make a wise commitment decision, based purely on the mathematics of optimal sequential search. It is about the decision process, not about any one relationship.
Sources
- Gilbert, J. R. and Mosteller, F. (1966). Recognizing the maximum of a sequence. Journal of the American Statistical Association, 61(313), 35-73.
- Todd, P. M. and Miller, G. F. (1999). From Pride and Prejudice to Persuasion: Satisficing in Mate Search. Simple Heuristics That Make Us Smart (pp. 287-308). Oxford University Press.