Fundamentalist
Agent type · fundamentalistRule
Buys when price < perceived fair value
What they represent
The slow, patient capital. Institutions. Value investors. Believes markets revert to truth.
01 · The Playground
1,000 agents.
Your rules.
Watch a market form.
This is a live simulation of a financial market — built from interacting agents, not stochastic assumptions. Price is not input; it is an output of how traders decide. Adjust the mix below. Inject a shock. Watch volatility clusters, bubbles, and crashes emerge on their own.
Emergence · The Playground
You're running a market.
500+ agents. Your rules.
Adjust the mix of trader types. Inject a shock. Watch prices emerge from their interactions — just like the real thing. No Gaussian assumptions. No closed-form pricing. Just agents.
● CALM
PRICE₹24500.00
VOL0.0%
STEP0
Fundamentalist
Chartist
Noise
Market Maker
Active trades
Preset Scenarios
Buy when price < fair value
Follow momentum
Random decisions
Based on Lux & Marchesi (1999). Runs entirely in your browser. Deterministic given seed.
Most traders see charts. We show you the minds that draw them.
02 · The Model
Lux-Marchesi, adapted.
Most financial models assume prices follow a clean stochastic process — Geometric Brownian Motion, Heston, jump-diffusion. These models give you closed-form answers but hide the ugly truth: markets are not noise-generating machines. They are people trading.
In 1999, Thomas Lux and Michele Marchesi published a different kind of market model in Nature. Instead of assuming a price process, they simulated the traders. Heterogeneous agents with simple rules. Different beliefs. Different reaction functions. And they found something remarkable: the price that emerged looked exactly like real markets — fat-tailed returns, volatility clustering, bubbles, crashes.
This playground is a reduced version of that idea. Four agent types. Deterministic rules. Everything runs in your browser at 60 fps. Same seed, same output — so you can share a configuration and we'll both see the same market.
03 · The Agents
Four archetypes. Every real market has them.
Chartist
Agent type · chartistRule
Follows momentum (short MA vs long MA)
What they represent
The trend follower. CTAs. Technical traders. Believes the tape tells the future.
Noise trader
Agent type · noiseRule
Random action each step
What they represent
The retail flow. News-chasers. Reactive. Doesn't believe anything consistently.
Market Maker
Agent type · mmRule
Mean-reverts inventory — buys when short, sells when long
What they represent
The liquidity provider. Profits on spread, not direction. Keeps the market functioning.
04 · What Emerges
Real market behaviors, unprogrammed.
Volatility clustering
Big moves come in bursts. Quiet periods last weeks. The simulation reproduces this without any volatility parameter.
Fat-tailed returns
Crashes happen more often than a Gaussian says. The distribution of price changes has heavy tails — like real markets.
Bubbles and crashes
Chartist dominance creates runaway momentum. Eventually the fundamentalists overwhelm — and the crash propagates in one step.
Regime switches
Shift the agent mix and the whole market behaves differently. No retraining, no parameters — just different participants.
05 · Reading list
The research this comes from.
Lux, T. & Marchesi, M. · 1999
Scaling and criticality in a stochastic multi-agent model of a financial market
Nature 397, 498–500
The foundational paper. Defines the chartist-vs-fundamentalist dynamics this playground uses.
Brock, W. A. & Hommes, C. H. · 1998
Heterogeneous beliefs and routes to chaos in a simple asset pricing model
Journal of Economic Dynamics and Control 22, 1235–1274
Adaptive belief switching — agents can change types based on performance.
Chiarella, C. & Iori, G. · 2002
A simulation analysis of the microstructure of double auction markets
Quantitative Finance 2, 346–353
Order-book-level extension with heterogeneous agents.
Farmer, J. D. & Joshi, S. · 2002
The price dynamics of common trading strategies
Journal of Economic Behavior & Organization 49, 149–171
Shows how price-impact effects combined with simple strategies produce fat-tailed returns.
Next