Play the Red Bead Experiment Simulator Online

The Red Bead Game is a hands-on simulation that demonstrates the difference between common-cause variation (random noise inside a stable system) and special-cause variation (signals worth investigating). Participants act as “willing workers” who use a paddle with 50 holes to scoop beads from a container that holds a mix of white beads and red beads. The red beads are defined as defects. No matter how hard workers try, the proportion of red beads they produce varies randomly around a stable average. The game is a vehicle for teaching that the system, not the worker, determines the result.

Bill Boller of Hewlett Packard created the Red Bead Game in 1982 as a gift for W. Edwards Deming. Deming then used the game in his famed four-day seminars over the last decade of his life and career. The game is documented in Four Days with Dr. Deming by William Latzko and David Saunders, and Mark Graban presents an updated version in Chapter 5 of Measures of Success.

“Red Bead Experiment” and “Red Bead Game” refer to the same exercise. Deming used both terms. The word “experiment” reflects the simulation’s connection to Deming’s teaching about statistical variation; the word “game” tends to feel less threatening to participants and is used more often in modern facilitation.

A facilitator plays the role of a manager at a fictional company, often called BaaS (Beads as a Service) in modern adaptations by Mark Graban. Six willing workers take turns scooping beads with a 50-hole paddle from a container holding white beads and red beads. Inspectors count the red beads on each paddle and a recorder writes the count down. Workers complete several rounds. The manager praises high performers, scolds low performers, and introduces interventions like targets, posters, and incentives. Across rounds, workers’ counts vary widely from one draw to the next, with no actual change in skill or effort. The game ends with a debrief that reveals the system was random the entire time.

The willing workers must use the paddle exactly as instructed (no picking individual beads), follow a deliberately over-detailed procedure, and accept whatever count the inspectors call out. Workers cannot question the process, change the procedure, or remove red beads from the container. The inspectors must count independently and agree on the number. The facilitator delivers all feedback, rewards, and punishments. These constraints are the point: workers have no actual influence over their outcomes, which mirrors how many real workplaces operate.

A typical setup uses around 4,000 white beads and 1,000 red beads, so 20 percent of the beads are red. The paddle has 50 holes. With a 20 percent defect rate, the long-run average is about 10 red beads per draw, but individual draws routinely range from roughly 4 to 16 red beads due to common-cause variation. Some adaptations use different totals; the principle stays the same.

“Willing worker” was Deming’s term for the participants who scoop beads with the paddle. The phrase emphasizes that people choose to come to work and that, in his words, “Workers will try to do a good job even when they know they cannot.” The label is a deliberate counterweight to the manager character’s habit of blaming individuals for results the system actually produced.

The Red Bead Game teaches eight related lessons: variation is built into every system; common-cause variation looks like trends but is not; ranking and rewarding people based on random results creates injustice; targets without a method are meaningless; exhortations, posters, and slogans do not change a system; workers cannot outperform the system that contains them; managers who react to every data point waste time and erode trust; and improvement requires changing the system, which is leadership’s job. As Deming put it, “The worker is not the problem. The system is the problem.”

The moral is that a stable system produces a predictable range of results, and reacting to individual data points within that range is at best wasted effort and at worst destructive. Praising the lucky, punishing the unlucky, and chasing the noise creates fear, distorts behavior, and leaves the underlying system untouched. Improvement requires changing the system, not pressuring the people inside it.

The workers cannot improve because the outcome is determined entirely by the proportion of red beads in the container and the random nature of the draw. Effort, attention, and skill do not change the underlying probability. The only ways to “improve” are to cheat (flicking red beads off the paddle, falsifying the count) or to change the system itself, which the workers are not permitted to do. This is a deliberate parallel to how many real systems work: people are held accountable for results they do not control.

Common-cause variation is the routine, predictable noise produced by a stable system operating as designed. Every result in the Red Bead Game is common-cause variation, even the unusually high and low draws. Special-cause variation, by contrast, is a signal that something has actually changed in the system. The whole point of a Process Behavior Chart is to tell these two apart, so leaders react to signals and ignore noise.

A full Red Bead Game session typically runs 60 to 90 minutes, including setup, four rounds of production, and a structured debrief. Faster versions can be completed in 30 to 45 minutes if rounds are abbreviated, though the lessons land hardest when participants get the chance to feel the manager’s escalating absurdity over multiple rounds.

Yes. An interactive web simulator built on Mark Graban’s adaptation runs the full four-day BaaS scenario in the browser, including the manager character’s escalation, the auto-generated Process Behavior Chart, and a debrief view. The simulator uses a hypergeometric draw from 5,000 beads (1,000 red, 4,000 white) and a 50-hole paddle, matching the standard physical setup. It is useful for solo learning, remote facilitation, and pre-class preparation before an in-person workshop.

A Process Behavior Chart, or PBC, plots the 24 paddle counts in time order and adds three reference lines: the average, an Upper Natural Process Limit, and a Lower Natural Process Limit. The limits are calculated from the data itself using the average moving range, not from arbitrary targets. In the Red Bead Game the average is typically near 10, the upper limit lands near 22 or 23, and the lower limit calculates to a small negative number, which becomes zero in practice because you cannot draw negative beads. The chart shows that every result was inside the limits and therefore part of routine variation.

Donald Wheeler describes three rules for finding signals on a Process Behavior Chart. Rule 1 is a single point outside the natural process limits. Rule 2 is eight or more consecutive points on the same side of the average. Rule 3 is three out of four consecutive points closer to a limit than to the average. If none of these rules fire, the system is predictable and the variation is noise. Reacting to noise as if it were signal is the core management failure the Red Bead Game illustrates.

Red beads are any defect that fluctuates because of system design rather than individual effort. Common examples include software bugs, customer support escalations, patient falls, surgical complications, manufacturing rework, and customer churn. When a Chief Medical Officer in one of Mark Graban’s workshops was asked the question, he answered that all of his patient safety measures were red beads. Recognizing the red beads in your own work is the first step toward managing them as a system instead of blaming the people closest to them.

Because most workplaces use the same playbook. Worker of the Day awards, performance probation, motivational posters, arbitrary targets, individual incentives, and layoffs of the bottom performers are all standard responses to variation that managers do not understand. The Red Bead Game compresses these moves into a single hour and strips away the cover story, so the absurdity becomes obvious. Participants laugh because they recognize their own organizations.

The most frequent criticisms are that the game is too simple, that real work involves skill and the simulation does not, and that the manager character is a caricature. Each criticism misses the point. The simulation is intentionally simple so that the only variable left is variation itself; if real work involves skill, then real work also involves variation on top of skill, which the game prepares people to see; and the manager character is a caricature because every behavior the character performs is documented in real workplaces, sometimes verbatim.

The Red Bead Game is a compact demonstration of three of Deming’s System of Profound Knowledge components: appreciation for a system, knowledge about variation, and psychology. It illustrates why ranking people, reacting to single data points, and using exhortations and targets fail. It also sets up the case for Process Behavior Charts, which Deming inherited from Walter Shewhart and which Donald Wheeler refined and popularized as a practical management tool.

The most useful single source is Chapter 5 of Measures of Success by Mark Graban, which walks through the full BaaS facilitation and ties the lessons directly to Process Behavior Charts. Deming’s own writing in Out of the Crisis and The New Economics provides the underlying theory. Four Days with Dr. Deming by Latzko and Saunders documents how Deming ran the experiment in his seminars. Donald Wheeler’s Understanding Variation is the canonical reference for the chart method that follows the game.