Pitcher or batter ?
Are home runs about the hitter or pitcher?
One way to address: use random effects model.
\[ \log \left(\frac{p}{1-p}\right) = \mu + \beta_i + \gamma_j \]
- \(\mu\) is the overall effect
- \(\beta_i\) is the effect due to the \(i\)th hitter
- \(\gamma_j\) is the effect due to the \(j\)th pitcher.
- \(\beta\) and \(\gamma\) are both assumed to be normally distributed:
- \(\beta_i \sim N(0,\sigma_h)\)
- \(\gamma_j \sim N(0,\sigma_p)\)