The Ballin with Beta Factor Playbook itemizes all asset pricing anomalies that I attempt to replicate in NBA spread markets. Some of them may end up being true systematic edges. Others may not; but just because I couldn’t get it to work doesn’t mean the edge doesn’t exist, it’s probably that I didn’t data-mine hard enough …JK 😉
| Factor | Win %* | t Stat (> 50%) | Betting Days | Cumulative Units Won** |
|---|---|---|---|---|
| Value | 51.0% | 1.77 | 2,253 | 43.22 |
| Momentum | 51.1% | 1.79 | 1,875 | 40.99 |
**Cumulative Units Won assumes 1 unit bet per day at +100 odds over the 15 NBA regular seasons between 2008 and 2023
Box-Score Stats in the Ballin with Beta Investable Universe
To translate factors from the quant finance world to the sports betting world, the Ballin with Beta approach starts by identifying common box-score stats that are known drivers of a team’s performance:
- Offensive Field Goal Percentage:
OFG\%\ =\ \frac{FGM\ +\ 0.5\cdot\left(3P\ FGM\right)}{FGA} - Defensive Field Goal Percentage:
DFG\%\ =\ \frac{FGM_{Oppt}\ +\ 0.5\cdot\left(3P\ FGM_{Oppt}\right)}{FGA_{Oppt}} - Offensive Turnover Rate:
OTO\%\ =\ \frac{TO}{FGA\ +\ 0.44\cdot FTA\ +\ TO} - Defensive Turnover Rate:
DTO\%\ =\ \frac{TO_{Oppt}}{FGA_{Oppt}\ +\ 0.44\cdot FTA_{Oppt}\ +\ TO_{Oppt}} - Offensive Rebound Rate:
ORB\%\ =\ \frac{ORB}{ORB\ +\ DRB_{Oppt}} - Defensive Rebound Rate:
DRB\%\ =\ \frac{DRB}{DRB\ +\ ORB_{Oppt}} - Offensive Free Throw Rate:
OFT\%\ =\ \frac{FTA}{FGA} - Defensive Free Throw Rate:
DFT\%\ =\ \frac{FTA_{Oppt}}{FGA_{Oppt}} - Offensive Assist Rate:
OAST\%\ =\ \frac{AST}{FGM} - Defensive Assist Rate:
DAST\%\ =\ \frac{AST_{Oppt}}{FGM_{Oppt}}
Each box-score stat is treated like a fund that bets on every NBA game every day, only betting on the team that screens more favorably on the factor being replicated. Given there are 10 “funds” in the investable universe, every day, we would have 10 PnLs, and each of the 10 funds bet on the same games simultaneously. So the “defensive rebound fund” would bet all games on the team with higher defensive rebounds per game, while the “offensive turnover fund” would bet all games on the team with lower offensive turnovers per game. These 10 “funds” are sorted into quantiles, based on a factor-specific sorting criteria, to determine which box-score stats are in the long leg and which box-score stats are in the short leg.
If over the course of multiple seasons, the return of the long/short portfolio vs. closing spread lines is positive and statistically significant, we can conclude that the factor being replicated is robust in sports betting markets, and may be a true edge.