Probability Concepts and Simulation

Population vs. Sample: A population consists of $N$ individuals, while a sample consists of $n$ individuals selected from the population.

Proportion Representation: Percentages are conventionally expressed in decimal form to directly reflect proportions.

$P(A) = \frac{\text{number of outcomes in event } A}{\text{total number of outcomes in sample space}}$

Frequentist Perspective:
- Probability is the proportion of times an outcome occurs in a very long series of repetitions, represented as a value between $0$ and $1$ .

Short-Run Misconceptions:
- Untrained intuition incorrectly expects randomness to be predictable in the short run.
- The "law of averages" (or "Gambler's Ruin") refers to the mistaken belief that chance outcomes must "even out" in the short run.

Simulation Definition: Using a model matching outcome counts and probabilities to imitate chance behavior and evaluate real-world results.
Trial Process:
- Assign numerical labels to outcomes (e.g., $1 = \text{Joey Logano}$ , $2 = \text{Kevin Harvick}$ , $3 = \text{Chase Elliott}$ , $4 = \text{Danica Patrick}$ , $5 = \text{Jimmie Johnson}$ ).
- Generate random integers from $1$ to $5$ until all outcomes appear, then record the required count.
Statistical Evidence Threshold:
- An observation occurring within the $5\%$ ( $0.05$ ) least likely outcomes of a simulation provides convincing statistical evidence.
- A probability of $0.0499$ is considered convincing evidence, whereas $0.0516$ is not.
NASCAR Cereal Example:
- Taking $23$ boxes to collect all $5$ driver cards yielded a simulated probability of $\approx 0/50 = 0$ across $50$ trials.
- Because this outcome is well below the $0.05$ threshold, it provides convincing evidence that the $5$ drivers' cards are not equally likely.