Galton board – Plinko probability
MathematicsProbabilityAges 13–14
How do you mainly use this simulation?
Use it with your class: students open it from a link or a QR code, with no account needed.
Drop balls through a triangular board of pegs: each time a ball hits a peg it bounces right with probability p, and after n rows of pegs it falls into a bin at the bottom. The Intro screen drops 1, 10 or 100 balls through a fixed 12-row board with p = 0.5. The Lab screen lets you change the number of peg rows (1–26) and the probability p with sliders, drop balls continuously, and compare the experimental mean/standard deviation with the theoretical binomial formula C(n,k)pᵏ(1−p)ⁿ⁻ᵏ. Based on the "Plinko Probability" simulation by PhET Interactive Simulations, University of Colorado Boulder (CC BY 4.0); rewritten. The Binomial screen sets n and p and shows the bars P(X = k) = C(n, k)pᵏ(1 − p)ⁿ⁻ᵏ or P(X ≤ k), the mean E(X) = np and variance V(X) = np(1 − p), the probability of an interval a ≤ X ≤ b, the counts of a Galton board beside them, and the normal curve N(np, np(1 − p)) with continuity correction when np and n(1 − p) are both at least 5.
Lesson: Experimental and theoretical probability
What it shows
On a Galton board each ball bounces right with probability p or left with probability 1 − p at every peg, independently of earlier bounces. After n rows the bin number k counts the right bounces, so k follows the binomial distribution P(X = k) = C(n, k)pᵏ(1 − p)ⁿ⁻ᵏ with mean np and variance np(1 − p). When np and n(1 − p) are both at least 5, the normal curve N(np, np(1 − p)) fits the bars well, and interval probabilities are approximated with the continuity correction a − 0.5 to b + 0.5.
How to use
On Intro, press Drop 1 ball, Drop 10 balls and Drop 100 balls and compare the bars with the theoretical curve. On Lab, change Peg rows n and Right-bounce probability p and tick Drop continuously. On Binomial, set Number of trials n, Probability of success p, From k = a and To k = b, switch Show between P(X = k) and P(X ≤ k), tick Normal curve, drop balls, then try the Task.
Parameters you can change
- Screen Intro, Lab, Binomial
- Number of peg rows (Intro, p = 0.5) 4–16 rows
- Starting number of peg rows (Lab) 1–26 rows
- Starting right-bounce probability p (Lab) 0–1
- Bin labels Count, Percent
- Show theoretical curve
- Lab: start dropping continuously
- Binomial: number of trials n 1–50
- Binomial: probability of success p 0–1
- Binomial: interval from k = a 0–50
- Binomial: interval to k = b 0–50
- Binomial: show P(X = k), P(X ≤ k)
- Binomial: normal curve
Questions to explore
- Why do the relative frequencies get closer to P(X = k) as more balls are dropped?
- How do the mean np and the spread of the bars change when p moves away from 0.5?
- For which values of n and p does the normal curve fit the binomial bars well, and why?