Sports Predictor includes two evaluation mechanisms: a test accuracy score printed during training, and evaluate_bet_teamwins() which translates a win probability into an actionable bet recommendation.
Test accuracy
After fitting the model, train_model_teamwins() prints accuracy on the held-out 30 % test split:
preds = model.predict(X_test)
print("Test Accuracy:", metrics.accuracy_score(y_test, preds))
# Example output: Test Accuracy: 0.XXX
Bet evaluation
evaluate_bet_teamwins(probability) takes the float returned by predict_game_teamwins() and returns one of three string labels:
def evaluate_bet_teamwins(probability):
if probability > 0.60:
return "Good Bet"
elif probability > 0.52:
return "Slight Edge"
return "Avoid"
Probability thresholds
| Condition | Label | Meaning |
|---|
probability > 0.60 | "Good Bet" | Strong model confidence — the edge over 50 % is meaningful |
probability > 0.52 | "Slight Edge" | Marginal advantage — proceed with caution |
probability <= 0.52 | "Avoid" | Not enough edge to justify a wager |
Test scenarios
The following cases reflect the test inputs defined in test_betting.py. Note that the test file references older function names (predict_game, evaluate_bet) that have since been renamed to predict_game_teamwins and evaluate_bet_teamwins. Use the current function names as shown below:
Strong Home Favorite
Slight Underdog
Even Match
Bad Bet Example
prob = predict_game_teamwins(
points_diff=10,
team_reb_roll=5,
opponent_reb_roll=3,
team_ast_roll=4,
opponent_ast_roll=2,
home=1
)
# Expected: probability well above 0.60 → "Good Bet"
print(evaluate_bet_teamwins(prob))
prob = predict_game_teamwins(
points_diff=-2,
team_reb_roll=4,
opponent_reb_roll=5,
team_ast_roll=3,
opponent_ast_roll=4,
home=0
)
# Expected: probability near or below 0.52 → "Slight Edge" or "Avoid"
print(evaluate_bet_teamwins(prob))
prob = predict_game_teamwins(
points_diff=0,
team_reb_roll=3,
opponent_reb_roll=3,
team_ast_roll=3,
opponent_ast_roll=3,
home=1
)
# Expected: probability near 0.50 → "Avoid"
print(evaluate_bet_teamwins(prob))
prob = predict_game_teamwins(
points_diff=-8,
team_reb_roll=2,
opponent_reb_roll=5,
team_ast_roll=1,
opponent_ast_roll=4,
home=0
)
# Expected: low probability → "Avoid"
print(evaluate_bet_teamwins(prob))
Use rolling stats from the most recent five games when building your inputs. Fresher averages reflect current form and produce more accurate probability estimates than season-long averages.