Phillies Win Title
From the International Herald Tribune:
The GCP assessment was stimulated by Brenda Dunne's suggestion:
We have a number of sporting events in the formal series, including some World Cup Soccer matches and previous World Series. For the GCP event in this case I added, as usual, some time before and after the nominal event, specifying 8 to 10:30pm local time. The result is an interesting graph with a steady positive slope though it's not significant. But at the end of the game there is a whopping downward spike for a few minutes. While acknowledging that interpretations of single events and graphs are unreliable, this is a really striking, unusual feature in my experience. In any case, the outcome is Chisquare 9139.4 on 9000 df for p = 0.149 and Z = 1.039. For an interesting extension, see the next graph.
The following figure explores a longer aftermath, responding to Brenda's remark about the excitement that followed the game. She said it seemed to involve the entire city of Philadelphia in a state of ecstasy, and was quite extraordinary. This segment shows a steeper positive slope from the time the game ended until midnight, and by itself shows what would be a significant deviation if this had been our formal prediction.
Extending the time still further, we include the continuing celebration through most of the next day in the following figure. The strong trend continues. Of course there are other things going on that are engaging people in wider circles around the world. Not least among these is the excitement of the election two days from now. So we can't ascribe the deviations from randomness in these figures only to the joy and celebration of the Phillies World Series win, but it is fair to suggest these emotions make a contribution.
It is important to keep in mind that we have only a tiny statistical effect, so that it is always hard to distinguish signal from noise. This means that every "success" might be largely driven by chance, and every "null" might include a real signal overwhelmed by noise. In the long run, a real effect can be identified only by patiently accumulating replications of similar analyses.