Pierce Egan's Boxiana: The Good Parts

I haven't posted about statistics in months and months. I figure I might as well use this for something.

I've been working for quite a while at reading the 1812 first volume of Pierce Egan's Boxiana; or Sketches of Ancient and Modern Pugilism. Volume one's subtitle is "From the Days of the Renowned Boughton and Slack, to the Championship of Cribb." It is an astounding tome, brought to my attention by the references made to it by A.J. Liebling in his fantastic collection "The Sweet Science." The full text of Boxiana is available to view on Google Books.

I have enjoyed it highly, and I've read about 120 of the books 420+ pages so far. I wanted a place I could put down some of the most delicious passages I read in the book, so that if you search for "Boxiana" online you can find this page to get a sample of what to expect. The volume is highly readable for something approaching its 200th birthday, and I'd say you could even enjoy it just as a sketch of the time even if you aren't particularly interested in boxing.

Now, on with the quotes. I'll use the page numbers on the Google Books edition so you can find where I got the passages from.

pp. 109-110

Mr. Brady, well known in the annals of pugilism, was the patron of Humphries, and prevented from attending the battle by business of a private nature; but who felt so anxiously for the issue of the contest, that his servant was ordered to witness the conflict, and convey to him the earliest intelligence. Humphries, immediately after the fight, like the heroes of old, wrote the following laconic epistle to Mr. Brady.

"Sir,

" I have done the Jew and am in good health.

"Richard Humphries."

p. 119

Warr again contended for victory with Dan Mendoza; and it should seem that Bill entertained an idea that there was a chance left of taking the shine out of the Jew; but a quarter of an hour sufficiently satisfied him it was grounded in error!

I'll post more as I find them, but here's two to start with.

Nintendo Wii Gameplay Stats

Earlier this month (weeks ago, I'm know), there were some very interesting posts on Kotaku.com about how long a person that buys or rents a Nintendo Wii game plays that particular game.

The posts can all be found under the Wii Stats Report tag on Kotaku, and they are making me think very creatively about ways this data could be used to change the way game marketing or game recommendation systems could possibly work, if this were a perfect world where this data was out in the public eye.

The page you probably want to start on would be this one:
The Ten Most Avidly-Played Wii Games in America (As Of July 1)
The games on this page are the ones that people are really getting their money's worth out of. The average person that owns Super Smash Bros. Brawl, which was purchased most likely for $50, has gotten 68 hours, 51 minutes of gameplay out of it! How many of any kind of multimedia item have you ever purchased at full MSRP and were able to get entertainment out of it at the cost of less than a dollar per hour?

Compare that to the average play time for Wii Fit, 18 hours and 18 minutes. For $90, each person that's buying that game is paying $5 an hour.

The wheels that spin in my head when reading these posts lead me to wish that there was some way to accomodate this kind of data into game reviews and recommendations. I imagine something like this:

"Compared to people in similar demographic groups to you and also to people overall, you only play your games about half as much as the average user. Also, you've put way more time into sports games than average and a lot less time in when you've purchased RPGs. Even though our review scores for Mario Super Sluggers and Fire Emblem are roughly the same, we can estimate that you'll get so much more for your money if you get Mario Super Sluggers! Based on average usage with adjustments for your preferences, we think you'll play the game for 30 hours. Not bad for $50, right?"

Of course, those are exactly the kind of things that will never exist because Nintendo and every other seller of everything everywhere secretly hopes that you don't play one game or use one thing too long so that you can come back and buy the next thing down the line. That, and because really the only person that this would help sell a game to is, well, me.

ULTIMATE FIGHTING CHAMPIONSHIP: Homegrown mixed martial arts league becomes major tourist draw

Found this article today about an independent study the UFC had done on the economic impact of the company on Las Vegas compared to other large Vegas events.

ULTIMATE FIGHTING CHAMPIONSHIP: Homegrown mixed martial arts league becomes major tourist draw

More than 4 million households order pay-per-view for six UFC events, data show

I find this passage particularly interesting:

The UFC generated $86.2 million in nongaming revenue for six events between
Feb. 2, 2008, and Jan. 31. Only the NASCAR UAW-DaimerChrysler 400, which
generated $134.3 million on March 2, 2008, ranked higher.

The UFC events attracted 80,087 people, with 56,435 of them coming to Las
Vegas for the event, numbers provided by Zuffa show.

The league's attendance numbers place the UFC behind, in order, National
Finals Rodeo, NASCAR, ACDelco NHRA Drag Races, the Aviation Nation Air Show,
NHRA SummitRacing.com race, the Professional Bull Riders National Championship
and the 10-day Cowboy Christmas Gift Show.

Those numbers show that the UFC has high ticket prices, we know that, but also that there are a lot of events that can be seen as "less famous" that are bringing in more people than the UFC was able to over a year period.

There's no information as to how much the average person pays to get into the PBR National Championship, for example. The UFC makes more money and certainly has better TV ratings than PBR, but are they charging too much for tickets or running buildings too small?

Also, where are they getting the $86.2 million value? The record UFC gates are in the neighborhood of $4-5 million, so with 6 events you're only talking about at most $30 million. Are they extrapolating money spent on hotels and casinos by attendees taking credit for that?

There's also some another dubious spot that really sticks out to me:

The study shows that just more than 4 million households ordered pay-per-view
for the six events in the study. That's a lot of exposure for Las Vegas, based
on the study's findings that each pay-per-view purchase has an average of eight
viewers. Based on that number, the six events garnered a total of 33.7 million
viewers.

Even if 8 people generally watch a show together, it's really pushing it to say that 4 million different households ordered the show when that is, I can guarantee, absolutely not the case. (Side note: I'm a nerd, the average audience when I buy a PPV is one.) Really, the number of "households" isn't going to be much more than the highest buyrate for a show, so if we are generous and grant them that 8 viewers along with only a small overlap, the true number of "viewers" relevent to this exposure statistic has to be more like 12-15 million.

I do like to see some of the interesting numbers regarding the make-up of the show's crowds, that's something nice to see that there hasn't really been a lot of data on before. It is, however, a bit misleading when you list the first bullet in the section as:

The UFC 94 was the primary reason 90 percent of the attendees traveled to Las
Vegas. Only 5 percent of visitors to Las Vegas come for a special event.

You can't sample people at a special event and ask them if they are coming for a special event when the sample you're comparing against is "all people."

I'd like to see the percent of visitors that came to the show on a plane vs. driving, that would probably give a better view of the national hold of the sport, then compare that to some of the larger sporting events like the Super Bowl.

Global Advertising: Consumers Trust Real Friends and Virtual Strangers the Most

Here's a post from today on Nielsen Wire about consumer trust in various sources of advertising:

Global Advertising: Consumers Trust Real Friends and Virtual Strangers the Most

Here's the chart in the article:




The most interesting thing I see here is the note at the bottom:

*E.g. 90 percent of respondents trusted "completely" or "somewhat" recommendations from people they know

Is that really what they gave people to choose as the top two categories? There are very few things I wouldn't pick "somewhat" for, and that has to apply to everyone because they surveyed 25,000 people and only 4 forms of advertising came back with less than 50% trust level!

I can understand the methodology of this in a way, because it's easier to get someone to pick between "none," "some," or "completely." Still, it's definitely deceiving to come back with a report saying "people trust brand websites just as much as they trust online reviews!" without giving a breakdown showing how much more people picked "completely" vs. "somewhat." Especially when they're also drawing the graph to say "and it's only a little less than they trust their best friend!"

Them's Numbers

Hello. I'm Matt. I'm interested in numbers, statistics, ratings, and things like that. Particularly related to business/marketing, but also ones tied to my various hobbies such as MMA, video games, TV, soccer, boxing, and comedy. When I find interesting reports on new marketing findings, research reports, or sometimes just neat things, I plan to post them on here. I may or may not offer editorial comments of my own. My goal is mostly to have one place to track the things I find interesting in the world of statistics and business, and to be able to share them with people that may be interested. So follow this, or not. Add it to your Google Reader, or not. Post comments, e-mail me, or not. Do what you want with your own life.