Mindlessly following Polymarket's pre-game win probability trades for NBA, can you make a guaranteed profit?

When trading NBA games on Polymarket, perhaps like many others, you’ve experienced this: before the game, one team’s win probability is clearly higher than the opponent’s, but in the fourth quarter, they suddenly collapse and get swept away (like the recent Hornets and Heat—those trades almost made me question my life).

Since everyone says Polymarket is a “truth machine,” does that mean if I blindly buy the team with the higher pre-game win probability, I can easily make money?

To verify this hypothesis, I backtested 1,096 regular-season NBA games from the 2025-26 season. The data’s truth is—

Blindly following the market won’t make you rich, but it won’t lose you much either; pre-game probabilities are already fully priced in.

Blindly buying according to the market, guaranteed profit

The backtest strategy used is very simple:

• Use the average pre-game probability over the first 3 minutes as the baseline
• $100 per trade
• Always buy the team with the higher win probability

Results:

• Total spent: $109,600; total returned: $107,545.20, a loss of $2,054
• ROI: -1.87%

This shows that Polymarket’s prices are already quite efficient; the market has fully priced in the teams’ win probabilities, leaving no arbitrage opportunities.

The ROI difference may come from transaction costs, emotional premiums, and other factors. Even blindly buying or going against the market can yield a 1.87% return.

The true value: teams and teams cannot be generalized

The above backtest was for the overall 1,000 games. I also broke it down from multiple angles to try to find parts that can break the market’s gravity:

• By week: random walk
• By probability: still a random walk. Betting on pre-game win probabilities of 50%, 60%, versus 70%, 80%, shows no difference in returns
• By team: here, significant differences appear

Some teams truly live up to market trust—

As long as the market believes they will 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 differ so much? Since I previously had limited knowledge about NBA teams, I had a hypothesis:

Are they the strongest or weakest teams, thus having high consistency in expectations?

But after verification, that’s not the case. Except for SAS (Spurs), the other four teams are only slightly above middle rank.

What about the best-performing teams? In fact, the market has already fully priced them in; blindly following and buying their average ROI is only 2.16%, with pre-game bets showing no inflated win probabilities.

• DET (Pistons): ROI 1%
• BOS (Celtics): ROI 4%
• NYK (Knicks): ROI 3%
• OKC (Thunder): ROI -2%
• DEN (Nuggets): ROI -5%

And the weakest teams?

Here, the divergence is extreme. These teams rarely get market pressure to win—e.g., the Nets (BKN) only have 7 wins with over 50% win probability, winning 5 of those, with an ROI of 21%; while the Pacers (IND) have only 8 games over 50%, winning 4, but an ROI of -20%. The sample size is too small to serve as a trading reference.

In other words, theoretically (just theory!), POR (Trail Blazers), PHI (76ers), SAS (Spurs), LAL (Lakers), and CHA (Hornets) are the ranges within which you can follow based on current data.