Jorge Espinosa wrote to suggest that we should consider the
effect on the GCP of the Philipino boxer Manny Pacquiao.
He literally shuts down the entire nation of the philipines whenever he
fights. There are zero crimes the day he fights, businesses shut down,
and come fight time, there can be upwards of 200 people gathered around
15 inch tv's, on the streets! EVERYONE sees it, and everyone has
their attention and positive vibes on their fellow countryman for at
least 1 hours or until the event ends.
Plus he's a different type of athlete; an actual
congressman, and gives back to his people as
opposed to other famous
boxers who just want to make a big paycheck to
inflate their
wallets. It's no wonder he shuts down his nation,
literally, when he performs.
I think this athlete is influential and charismatic enough to
have drawn enough worldwide buzz in his past bouts.. The
charisma thing, is an understatement, I mean, I can't think of
any active world-class athletes who are currently congressmen
in their country.
His last bout was on May 7th, 2011. The event started at
9:00pm ET with two opening undecard bouts leading up to the
main event. He actually stepped into the ring between
11:15-11:45pm ET. I don't have exact times at the moment (I can
find out though, just need a day). The other bout before this
one was on Nov. 13th 2010 with pretty much the same time and
sequence of undercard bouts/main-event. and before that, March
13th 2010, and before that, Nov. 14th 2009. and one in
particualar vs Oscar de la Hoya, (his "break-through fight" you
could call it) on Dec 6, 2008.
This sounded promising, so I set a consistent GCP event for
each of 5 fights, using 11 pm to 2 am ET to get a bit of the
pre-fight flavor and then the bout and some post bout time.
The accumulation is shown in the first figure, with the
individual bouts following.
The individual graphs show how this accumulation comes
about. Four of the five fights correspond to a positive
deviation in the GCP data, and two of those are quite
significant, with odds less than 1 in 100. The fight dates
are shown in the title, in the format YYMMDD.
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.
|