Why is “one chance in ten” not the same as “it will happen once in ten tries”?
Quick answer
“One chance in ten” means each qualifying try has a 10% probability of success, if the chance stays the same. It does not promise exactly one success in every group of ten tries. Even when tries are independent, a group of ten can have no successes, one success, or several. The probability applies to each try; the number of successes in a particular batch remains uncertain.
The question, the details, the next step.
If something has “one chance in ten,” doesn’t that mean it should happen once in every ten tries?
No. It means the chance is one in ten for each individual try, not that a batch of ten is guaranteed to contain exactly one success. Think of a hypothetical spinner with ten equally likely outcomes, where landing on one chosen outcome counts as success. A spin has a one-in-ten chance of success; after ten spins, you might see zero successes, one, or more.
The distinction is between a probability for a try and a count of what happened across a batch. A useful comparison is a fair coin: the chance of heads on one toss is one-half, but a short run of tosses need not contain exactly half heads. [S1] The same idea applies here. A probability describes uncertainty; it is not a schedule.
What assumptions are behind that answer? Does it matter whether the tries affect one another?
Yes, it matters. The simplest calculation assumes a fixed number of tries, the same success probability on each try, and independent tries—meaning the result of one try does not change the probability on another. Those are the conditions for a binomial experiment. [S2]
For example, suppose a hypothetical event has a 10% chance on each of ten independent tries. Under those assumptions, the chance of no successes is 0.9 × 0.9 × … × 0.9 (ten factors), or 0.9¹⁰, about 35%. So even zero successes is a possible batch outcome. If trying again changes the setup, or the chance varies from try to try, this calculation may not fit. The “one in ten” figure alone does not tell us whether those assumptions hold.
Can you show why even exactly one success in ten is not guaranteed?
Using the same hypothetical assumptions, the probability of exactly one success is 10 × 0.1 × 0.9⁹, about 39%. The factor of 10 accounts for which of the ten tries is the successful one; the other nine must be failures. This follows the binomial approach for finding the probability of a specified number of successes in a fixed set of trials. [S2]
That result is less than 100%, so “exactly one” is not guaranteed. Other outcomes also have probability: there may be no successes, or two or more. These figures illustrate the math for the stated assumptions; they do not establish the real probability of any particular event. If the quoted one-in-ten chance is only an estimate, or the tries are related, the example may not describe the real situation.
How should I explain this simply to someone else without making it sound as if the estimate is a promise?
You could say: “A one-in-ten chance means each try has a 10% chance, assuming the probability stays the same. Ten tries don’t guarantee one success; they could produce none, one, or several. The result depends on how the tries work and is uncertain.”
For a specific situation, state the assumptions next to the number. Say whether the probability applies to each try, whether it stays the same, and whether tries are independent. If those details are unknown, present the calculation only as an illustration, not a prediction or promise.
A practical checklist
Define what counts as one try and what counts as success.
Check whether the quoted probability applies to each try or to a whole batch.
State whether the chance is assumed to stay the same across tries.
Ask whether one try can affect the next; if so, the independent-tries calculation may not apply.
Describe batch outcomes as possibilities, not guaranteed counts.
Label calculations as hypothetical when the assumptions have not been confirmed.
A few more useful answers
Does a one-in-ten chance mean ten tries average one success?
Under the same-probability assumptions, the expected number across ten tries is one, but that is an average over many comparable batches, not a promise for any particular batch. [S2]
What if the chance changes after each try?
Then a single fixed per-try probability may not describe the situation, and the simple ten-try calculation may not apply. You would need a model that accounts for how the chances change.
Does ten tries make success certain?
No. Under the example’s assumptions, ten tries still have a positive chance of producing no successes. A per-try probability is not a guarantee about the batch.
Sources
References used for this guide.
- [S1] Definitions of Statistics, Probability, and Key Terms — Introductory Statistics openstax.org · Checked
- [S2] Key Terms — Introductory Statistics 2e openstax.org · Checked