## Quadrant 1

Predictable Gaussian World (Simplicity and Normal Distribution)

### Definition

Welcome to the Gaussian world of normal distributions, where there is a high degree of predictability and a low probability of fat tail events. Prediction errors are tiny and the probability of forecasting is high.

### Criteria

This is a Simple world with Thin tails. Predictions lead to results largely as expected. The payoffs are Simple and the distribution of events is Normal.

### Examples

- There are many businesses, investments, games and events where the outcome is binary; no other outcome but one of two. Elections, coin tosses and simple games of chance are of the win/lose variety.
- Distributions with small ranges are also part of Q1: with height, weight or age of people, one outlier cannot materially change the distribution (one person living to 110 will not change the average age of a city).

### Risk Management Tools

- Distributions are normal, but often with a slight “house” skew, so use Value-at-Risk (VaR) models.
- Define the statistics (your probability of winning or being right); NY state slot machines pay out 93.75%, Ohio just 90.3%.
- Define your maximum loss tolerance and stick to it.
- Normally distributed games of chance will usually generate a time series return that, when charted, resembles a sine wave; an oscillation above and below a horizontal zero line. If you’re paying a house or transaction charge, your sine wave will be around a downward sloping line.
- The only way to ‘win’ with games of chance is to recognize the mean-reverting nature of returns and to exit on the rising portion of the sine wave. And better to exit early than late.

## Quadrant 1

Predictable Gaussian World (Simplicity and Normal Distribution)

### Definition

Welcome to the Gaussian world of normal distributions, where there is a high degree of predictability and a low probability of fat tail events. Prediction errors are tiny and the probability of forecasting is high.

### Criteria

This is a Simple world with Thin tails. Predictions lead to results largely as expected. The payoffs are Simple and the distribution of events is Normal.

### Examples

- There are many businesses or events where the outcome is binary; no other outcome but one of two. Elections, coin tosses and simple games of chance are of the win/lose variety.
- Distributions with small ranges are also part of Q1: with height, weight or age of people, one outlier cannot materially change the distribution (one person living to 110 will not change the average age of a city).

### Risk Management Tools

- Distributions are normal, but often with a slight “house” skew, so use Value-at-Risk (VaR) models.
- Define the statistics (your probability of winning or being right); NY state slot machines pay out 93.75%, Ohio just 90.3%.
- Define your maximum loss tolerance and stick to it.
- Normal distributions, when charted, often appear roughly as a sine wave; an oscillation above and below a horizontal zero line. If you’re paying a house or transaction charge, your sine wave will be around a downward sloping line.
- The only way to ‘win’ with games of chance is to recognize the mean-reverting nature of returns and to exit on the rising portion of the sine wave. And better to exit early than late.

**Risk Quadrants**