Did you know? A result above 1 is a useful alarm: it points to the exact place where the probability model needs checking.
No. In standard probability, an event’s probability must be between 0 and 1 inclusive, or between 0% and 100%. A value greater than 1 signals a calculation, counting or interpretation error—not an exceptionally likely event.
For equally likely outcomes, probability is favourable outcomes divided by all possible outcomes. The favourable set cannot contain more outcomes than the sample space it belongs to, so the fraction cannot exceed 1.
Try this now: if an answer is 7/6, list the six possible outcomes and circle the favourable ones. A seventh favourable outcome cannot be added unless the sample space also changes. This exposes double-counting faster than simply redoing the division.
Choose the task you need
Anchor the probability scale
OpenStax states the probability rule as 0 ≤ P(A) ≤ 1. Zero represents an impossible event in the model, one represents a certain event, and values between them represent degrees of likelihood.
| Probability | Percentage | Interpretation |
|---|---|---|
| 0 | 0% | Impossible in the stated model |
| 0.25 | 25% | One chance in four under equal-likelihood assumptions |
| 0.5 | 50% | Even chance |
| 1 | 100% | Certain in the stated model |
| 1.2 | 120% | Not a valid event probability |
The words impossible and certain belong to the model and conditions. A probability model can be incomplete or based on assumptions, but its assigned probabilities still obey the 0-to-1 scale.
Find why the fraction exceeded 1
In a simple equally likely model, P(A) = number of favourable outcomes ÷ number of outcomes in the sample space. A numerator larger than its own denominator usually reveals that outcomes were counted twice, the denominator describes the wrong experiment, or non-equally-likely cases were treated as equal.
| Work shown | What went wrong | Repair |
|---|---|---|
| 7 favourable outcomes out of 6 die faces | At least one face was double-counted | List {1,2,3,4,5,6} once |
| 8 red cards ÷ 4 suits | Mixes different counting units | Use red cards ÷ all cards |
| P(A or B)=P(A)+P(B) with overlap | Overlap counted twice | Subtract P(A and B) |
| Favourable routes ÷ named destinations | Routes and destinations are not equiprobable outcomes | Define the sample space consistently |
A neat fraction bar cannot rescue mismatched counting units. Name what the numerator counts and what the denominator counts before calculating.
Distinguish probability from numbers that may exceed 1
Several nearby ideas can legitimately be greater than 1, which is why a learner may carry the wrong scale into probability.
| Quantity | Example | Can exceed 1? |
|---|---|---|
| Probability | 0.7 | No |
| Odds in favour | 3:1 | The first term may exceed 1; this is a ratio, not a probability |
| Expected number | 2.4 goals | Yes |
| Percentage increase | 150% increase | Yes |
| Relative frequency count | 14 successes | Yes; divide by total for a proportion |
| Probability density value | A curve height above 1 | In advanced continuous models, yes; area, not height, is probability |
At Primary and early secondary level, the most useful comparison is probability versus odds, counts and percentages. Probability density is an advanced caveat, not a reason to accept P(A) > 1.
Diagnose the exact reasoning error
| Observed answer | Likely gap | Useful prompt |
|---|---|---|
| 7/6 | Double-counting or wrong sample space | Can one outcome appear twice in the list? |
| 120% probability | Confuses percentage change with chance | Could the event happen more than every time? |
| 3:1 written as P=3 | Confuses odds ratio with probability | What fraction is 3 favourable out of 4 total parts? |
| Adds overlapping events directly | Does not see the intersection | Which outcomes satisfy both events? |
Diagnosis-first tuition chooses the next task from the error. A sample-space list helps double-counting; a two-way table or Venn diagram helps overlap; converting odds 3:1 to probability 3/4 repairs a different misconception.
Check and explain the final answer
- Confirm the value is between 0 and 1 inclusive.
- Convert to a percentage and confirm it is between 0% and 100%.
- Label numerator and denominator with the same type of outcome.
- Check whether outcomes are equally likely before using simple counting.
- For “or” events, test for overlap before adding.
A clear written explanation is: “No. Probability cannot exceed 1. My value 7/6 shows that I counted more favourable cases than exist in the sample space, so I must rebuild the outcome list.”
Current pathway note, checked 12 October 2026: Posting Groups 1, 2 and 3 support admission to secondary schools; they are not subject levels. Individual secondary subjects may be taken at G1, G2 or G3 according to the student’s programme and the subject available. See MOE’s Full Subject-Based Banding overview. The Singapore-Cambridge Secondary Education Certificate (SEC) begins in 2027. The applicable syllabus depends on the subject, level, examination year and candidate type; not every subject is available at every level.
A calm next step
If a child obtains 7/6 or 120% and cannot see why it is impossible, bring the full sample-space working. The useful diagnosis is whether the gap concerns double-counting, mismatched units, overlapping events or confusing odds with probability.
eduKateSG’s verified 3-pax positioning allows close attention to the exact step where an answer begins to drift. There is no grade guarantee; the useful question is whether the teaching response matches the learner’s observed error.
Ask eduKateSG about a 3-pax learning conversation. Bring one recent piece of work so the discussion can begin with evidence.
Sources and checking notes
- OpenStax Introductory Statistics, Terminology
- OpenStax Introductory Statistics 2e, Probability Key Terms
- OpenStax College Algebra 2e, Probability
- MOE, Full Subject-Based Banding
- SEAB, Singapore-Cambridge Secondary Education Certificate
Teaching examples and diagnostic routines in this guide are original eduKateSG illustrations. Policy links and supporting sources were checked on 12 October 2026.