[shinsengumi]

| (Predicted Stanford score - Predicted Opponent score) - (Actual Stanford score - Actual Opponent score) |

BC

I believe the formula used was a sum of errors:
| (Predicted Stanford Score - Actual Stanford Score) | + | (Predicted Opponent score - Actual Opponent Score) |

There are pros and cons to both.

[shinsengumi]

| (Predicted Stanford score - Predicted Opponent score) - (Actual Stanford score - Actual Opponent score) |

BC

I believe the formula used was a sum of errors:
| (Predicted Stanford Score - Actual Stanford Score) | + | (Predicted Opponent score - Actual Opponent Score) |

There are pros and cons to both.