Table VI.
Shocks from Emotionally Salient Games.
| Poisson Regression Intimate Partner Violence, Male on Female, at Home |
||||
|---|---|---|---|---|
| Game Type = Still in Playoff Contention |
Game Type = Traditional Rivals |
Game Type = Sacks≥4, Turnovers≥4, or Penalties>80 yds |
Game Type = Highly Salient: (1) & [(2) or (3)] |
|
| (1) | (2) | (3) | (4) | |
| More Salient Games (Game Type = 1) | ||||
| (a) Loss × Predicted Win (Upset Loss) | .126 (.034) | .197 (.046) | .151 (.048) | .172 (.045) |
| Loss × Predicted Close (Close Loss) | .054 (.031) | .011 (.053) | .027 (.038) | .082 (.046) |
| Win × Predicted Loss (Upset Win) | .027 (.048) | .156 (.080) | .083 (.040) | .063 (.059) |
| Predicted Win | −.021 (.028) | −.042 (.036) | −.055 (.035) | −.042 (.029) |
| Predicted Close | −.040 (.034) | −.021 (.051) | .019 (.038) | −.068 (.044) |
| Predicted Loss | −.023 (.033) | −.042 (.055) | −.024 (.026) | .010 (.038) |
| Less Salient Games (Game Type = 0) | ||||
| (b) Loss × Predicted Win (Upset Loss) | −.016 (.080) | .080 (.034) | .070 (.037) | .028 (.041) |
| Loss × Predicted Close (Close Loss) | −.003 (.030) | .035 (.026) | .042 (.034) | .018 (.028) |
| Win × Predicted Loss (Upset Win) | .002 (.039) | −.011 (.030) | −.024 (.033) | −.004 (.027) |
| Predicted Win | −.013 (.055) | −.014 (.027) | −.009 (.028) | −.010 (.030) |
| Predicted Close | .032 (.032) | −.012 (.029) | −.012 (.032) | .004 (.029) |
| Predicted Loss | −.008 (.028) | −.012 (.020) | .006 (.027) | −.023 (.021) |
| Non-game Day | --- | --- | --- | --- |
| Salience Test: | ||||
| p-value for row (a) = row (b) | .11 | .01 | .17 | .01 |
| Number of Agencies | 764 | 764 | 764 | 764 |
| Observations | 79,386 | 79,386 | 79,386 | 79,386 |
Notes: Standard errors in parentheses, clustered by team×season. Regressions include agency fixed effects, season dummies, week of season dummies, and the holiday and weather variables described in the note to Table 4. Estimated models are comparable to the baseline model in column 3 of Table 4. See notes to Table 4 for details. Each column is a single regression which allows for separate coefficients by game type. Still in playoff contention indicates that a team has a greater than 10% chance of making the playoffs given their current win-loss record, based on the probability that any NFL team made the playoffs with such a win-loss record between 1995 and 2006. Traditional rivals indicates a game between traditional rivals, as defined by “Rivalries in the National Football League” on Wikipedia.