How to use Probability Distribution Explorer
- Choose Binomial, Poisson or Normal and enter that model’s parameters. The other distributions’ fields are ignored.
- Choose Exactly x, At most x or At least x, then enter Evaluation x. Exactly x means density for Normal, not probability at one exact value.
- Set Plot start and Plot end separately from the evaluated x. Discrete endpoints are whole numbers; Binomial plot end cannot exceed positive n, and discrete spans are at most 200 steps.
- Calculate and compare the requested value with the cumulative table and distribution-plot.svg. A short plotted range can omit tails even when the evaluated probability is valid.
Example: Probability Distribution Explorer
For 10 independent trials with success probability 0.5, calculate the probability of exactly five successes and the cumulative probability through five.
Options
- Distribution assumptions
- Binomial uses a fixed number of independent trials with one success probability. Poisson uses the entered rate; normal uses a mean and positive deviation.
- Evaluation versus plot
- Choose the probability operation separately from the displayed plot range. A short plot range can leave probability mass in unseen tails.
Supported inputs and limits
- Binomial n is a whole number from 0 to 1,000, p is 0 to 1, and Poisson mean λ is 0 to 1,000.
- Normal SD is 0.000001 to 1,000,000.
- Evaluation x and plot bounds are within ±1,000,000; discrete plots need nonnegative integer starts and span at most 200 steps.
- Poisson plot end is at most 5,000.
- Normal plots use 101 samples; discrete plots use up to 201.
More details
Chosen models require appropriate independent-trial, constant-rate or normal assumptions; the tool does not fit a distribution to data.
Normal output is density at x or a tail area, while discrete point output is probability mass.
Discrete upper tails include x, so upper and lower inclusive tails overlap at that point.
Finite floating-point values can underflow to zero.
The plot can omit tails.
Where your input is processed
This tool does not upload your inputs. See the limits for anything saved on your device.
Density at a point is not a point probability
For a continuous normal model, the displayed height is density. Interval or tail probabilities describe areas under that density. Binomial and Poisson outputs use discrete masses at whole values. Keep the chosen model and operation with the result so those meanings do not get mixed.
Questions about Probability Distribution Explorer
Is a normal density the probability of one exact value?
No. A continuous distribution assigns zero probability to one exact value; probabilities belong to intervals.
Can the plot show every Poisson outcome?
No. Poisson outcomes are unbounded. The page shows a finite plotted range and states that limit.
Does the page select a distribution for my data?
No. Choose a model whose assumptions match the question you are studying.