Can Blindly Following Polymarket's Pre-Game Win Probability for NBA Trading Guarantee Steady Profits?
- Core Insight: Backtesting analysis of NBA game prediction markets on Polymarket reveals that market pricing is already quite efficient. Blindly following the side with a higher pre-game win probability does not yield excess returns. However, when segmented by specific teams, some teams show significant positive return differentials.
- Key Findings:
- Overall backtesting shows that across 1,096 games, consistently buying the side with a higher pre-game win probability resulted in a final ROI of -1.87%, indicating the market is fully priced with no arbitrage opportunities.
- Strategies segmented by probability intervals or by week all exhibited random walk characteristics in returns, with no systematic patterns discovered.
- When segmented by team, the follow-the-probability strategy for some teams (e.g., Trail Blazers, 76ers, Spurs, Lakers, Hornets) showed significantly positive ROI (9%-19%), indicating performance above the market average.
- For top-performing teams (e.g., Celtics, Nuggets), market pricing is already highly efficient, with the follow strategy's ROI close to zero or negative, validating market effectiveness.
- For bottom-performing teams, the sample size is too small, and their return data (e.g., Nets ROI 21%, Pacers ROI -20%) exhibits extreme volatility, lacking stable reference value for trading.
Trading NBA games on Polymarket – perhaps you, like many others, have had this experience: before the game, you see one team's win probability is significantly higher than their opponent's, only for them to suddenly collapse in the fourth quarter and get swept away (like the recent Hornets vs. Heat game; that bet made me question my life choices).
Since everyone says Polymarket is a "truth machine," does that mean I can just mindlessly buy the team with the higher pre-game win probability and easily make money?
To test this hypothesis, I backtested the 1,096 regular-season games of the NBA 2025-26 season. The data reveals the truth –
Mindlessly following the market won't make you money, but you won't lose much either. Pre-game probabilities are already fully priced in.
Mindlessly Following the Market Guarantees Losses
The backtesting strategy used was very simple:
- Used the average probability from 3 minutes before the game as the baseline.
- Traded $100 per game.
- Always bought the side with the "higher win probability."
Results:
- Total cost: $109,600. Total returns: $107,545.2. Net loss: $2,054.
- ROI: -1.87%.
This indicates that Polymarket's prices are quite efficient. The market has already fully priced in the teams' win probabilities, leaving no "arbitrage" opportunity.
The ROI difference likely stems from other dimensions like transaction costs or emotional premiums. If you insist on "mindlessly buying," you'd be better off betting against the market, which could yield a 1.87% profit.
The Real Value: Not All Teams Are Created Equal
The above backtest was for the entire set of one thousand games. I further broke down the data from multiple angles, trying to find segments that could break free from market gravity:
- By week: Random walk.
- By probability: Still a random walk. Betting on pre-game win probabilities of 50%, 60% showed no difference in returns compared to 70%, 80%.
- By team: Here, significant differences emerged.
Some teams simply live up to the market's trust –
When the market thinks they'll win, they are more likely to actually win.
- POR (Trail Blazers): ROI 19%
- PHI (76ers): ROI 14%
- SAS (Spurs): ROI 12%
- LAL (Lakers): ROI 11%
- CHA (Hornets): ROI 9%
Why do these teams show such differences? As the author previously had limited knowledge of NBA teams, a hypothesis emerged:
Are they the strongest or weakest teams, leading to high consensus in expectations?
However, verification showed this wasn't the case. Except for SAS (Spurs), the other four teams were only ranked in the middle to slightly above-average positions.
What about the teams with the best records? The market has already priced them in fully. Mindlessly buying them yields an average ROI of only 2.16%. The pre-game betting win probability contains no "water."
- DET (Pistons): ROI 1%
- BOS (Celtics): ROI 4%
- NYK (Knicks): ROI 3%
- OKC (Thunder): ROI -2%
- DEN (Nuggets): ROI -5%
What about the weakest teams?
Here, the results are extremely polarized. For such teams, there are hardly any games where the market favors them to win. For example, the Nets (BKN) had only 7 games with a win probability >50%, winning 5, resulting in a high ROI of 21%. Meanwhile, the Pacers (IND) had 8 games >50%, winning 4, but their ROI was -20%. The sample size is too small to serve as a trading reference.
In other words, theoretically (only theoretically!), POR (Trail Blazers), PHI (76ers), SAS (Spurs), LAL (Lakers), and CHA (Hornets) are the teams the existing data suggests you could follow.
