Second-Half Adjustment Rating

Overview

This pillar analyzes a college basketball team's ability to adjust its performance from the first to the second half. It identifies elite coaching staffs and resilient teams, providing a crucial edge for live and second-half betting markets.

What It Does

The Second-Half Adjustment Rating (SHAR) calculates a team's net rating improvement after halftime over a series of games. It compares offensive and defensive efficiency in both halves, adjusting for opponent quality to isolate true coaching and performance shifts. This reveals teams that consistently get better as the game progresses versus those who fade under pressure.

Why It Matters

Most betting models use full-game stats, creating a blind spot for in-game dynamics. This pillar exploits that by pinpointing teams that systematically outperform second-half expectations, offering a significant advantage in live betting where lines are constantly adjusting.

How It Works

The model ingests half-by-half box score and play-by-play data for every game. It first calculates the net efficiency rating for each half. Then, it computes the difference between the second and first half ratings. This raw adjustment score is then normalized based on the opponent's strength and averaged over a 10-game rolling window to generate the final rating.

Methodology

The core formula is SHAR = Average[(NetRtg_2H - NetRtg_1H)_adj] over the last 10 games. Net Rating (NetRtg) is calculated as (Points Scored per 100 Possessions) - (Points Allowed per 100 Possessions). The opponent adjustment (_adj) normalizes the score based on the opponent's season-long KenPom rating to contextualize performance.

Edge & Advantage

This pillar provides a predictive edge by quantifying a coach's in-game tactical adjustments, a factor that simple statistical models completely ignore.

Key Indicators

• Net Rating Swing

  high

  The opponent-adjusted change in points per 100 possessions from the first half to the second.

• Comeback Frequency

  medium

  The percentage of games a team wins when trailing by 5 or more points at halftime.

• Blown Lead Rate

  medium

  The percentage of games a team loses when leading by 5 or more points at halftime.

Data Sources

• NCAA Play-by-Play Data

  Provides raw possession and scoring data required to calculate half-specific efficiency ratings.

• KenPom Advanced Stats

  Source for opponent-strength ratings used to adjust the raw performance data.

Example Questions This Pillar Answers

• → Will Duke cover the -4.5 second-half spread against UNC?
• → Will a team trailing by 10 at halftime make a comeback and win the game?
• → Which team is a better bet on the live moneyline at halftime?

Tags