NBA Postseason Analysis: Expected Points Difference for 49 Wins

Intro

A team with 49 wins has never missed the playoffs. This analysis calculates the expected points difference (\(\text{PTSdiff}\)) required for a team to achieve 49 wins and secure a postseason spot, using the linear regression model WinsReg described in Norge Pena Perez’s RPubs publication.

---

R Calculation

The following R code calculates the required points difference dynamically.

# Define target wins
target_wins <- 49

if (file.exists("NBA_train.csv")) {
  # Load the dataset
  NBA <- read.csv("NBA_train.csv")
  
  # Compute points difference
  NBA$PTSdiff <- NBA$PTS - NBA$oppPTS
  
  # Fit the regression model
  WinsReg <- lm(W ~ PTSdiff, data = NBA)
  
  # Extract exact coefficients
  intercept <- coef(WinsReg)[1]
  slope <- coef(WinsReg)[2]
  
  # Calculate points difference
  pts_diff <- (target_wins - intercept) / slope
  
  cat("Dataset found. Expected points difference (trained model):", round(pts_diff, 2), "\n")
} else {
  # Fallback to RPub coefficients
  intercept <- 41.00
  slope <- 0.03259
  
  # Calculate points difference
  pts_diff <- (target_wins - intercept) / slope
  
  cat("Dataset not found. Expected points difference (RPub coefficients):", round(pts_diff, 2), "\n")
}

## Dataset not found. Expected points difference (RPub coefficients): 245.47