Last month, WWE filed their annual pre-WrestleMania lawsuit to protect against “trademark infringement, counterfeiting and dilution under the Lanham Act”. On my blog, I noted that Exhibit 3 of the filing included a detailed schedule of WWE’s touring plans covering March 2015 through March 2016.

In addition, the calendar entries included other interesting datapoints such as WWE Arena codes and related Arena Capacities, travel times and distances between cities, % number (assumed to be the percent of tickets sold during the last time WWE toured each city ) and a letter grade (A/B+/B/C) assigned to each location (believed to represent the “drawing power” of the city – basically tiering the expected number of tickets sold).

The percentage data, in particular, provides an interesting opportunity to see if we can correlate House Show attendance to specific headliners. (It’s an exercise I’ve done in the past.) Previously, I’ve used attendance numbers gleamed from the Wrestling Observer Newsletter or websites such as WrestlingData.com. This is the first time that I’ve been able to utilize internal WWE figures. I wonder whether that will that generate different information.

First of all, there’s the dataset. In total, the WWE schedule that was included in the lawsuit provide insight on 386 shows scheduled between March 1, 2015 (Raw at “NY100”, Nassau Coliseum, in Uniondale, NY – capacity 12,157 – start time of 7 PM) and March 28, 2016 (Raw at “DC005”, Verizon Center, in Washington, DC – capacity 14,296 – start time of 7:30 PM). In addition, WWE lists a “percentage number” for 214 shows (some of them representing the same venue/show more than once) ranging from 23% (MI020 – East Lansing, Michigan’s “Breslin Student Event Center” – which drew an estimated 2,626 fans on Friday, September 13, 2013) to 100% (TX066 – Hildago, Texas’ “State Farm Arena” – which drew an estimated 6,545 fans on Saturday, April 12, 2014). Shows with percentages stretch back from February 12, 2011 (Raw brand House Show at the “Save on Foods Memorial Centre” in Victoria, BC drawing 28%, estimated at 2,230 fans) to as recent as the February 16, 2015 (Monday Night Raw at “Amway Center” in Orlando, FL drawing 100%, estimated at 11,867 fans).

For this analysis, I looked at specific subset of data: (a) recent events, (b) house shows.

Why? Television and PPV tapings often sell on the overall popularity of the WWE brand. Wrestling fans have been trained to recognize that important things only happen when cameras are around. Raw, SmackDown and PPVs are much less driven by the promised appearance of any given top superstar. Furthermore, WWE rarely competes with themselves on TV days. They have all of their top talent at their disposal (barring injuries, suspensions or other issues).

House shows are a very different story. WWE usually runs two tours during the weekends (Friday-Sunday). This may take the form of separate overseas crews or two groups touring different parts of North America. Usually, both teams converge at either Monday’s Raw (or Sunday’s pay-per-view event). House shows provide an interesting study. With the top stars for each house show usually advertised, will fans decide to attend the house shows based on who is coming to their town?

I focus on the recent house shows (March 2014-February 2015) for a few reasons. First of all, limiting the timeframe removes “stars” that haven’t been with the company for years (sorry Mason Ryan & Ted DiBiase!). It also focuses on the cities that WWE often visits. If a town hasn’t been visited for years there’s usually a reason. It may say something about how poorly things went last time (drawing only 43% of the building, estimated at a 3,600 fans, for a Daytona Beach house show at the “Ocean Center” on 6/22/2013). Or it may be an indication of how remote the location is (for instance, the last tour of Alaska appears to have been January 2013).

Compare that to a popular locations like the Barclay’s Center in Brooklyn which WWE will visit five times over the course of 13 months (house shows in March 2015 & March 2016, SummerSlam 2015 and Raw tapings for August 2015 and December 2015). Another example would be the five trips to the Allstate Arena in Rosemont, IL (house show in December 2015, Extreme Rules 2015, Raw tapings in July 2015 & October 2015 & March 2016). Look at the Western New York loop with cities such as Rochester (March 2015/September 2015/March 2016 house shows), Syracuse (May 2015/September 2015/March 2016 house shows) and Buffalo (March 2016 house show and SmackDown tapings in March 2015 & June 2015 and Raw taping in September 2015). Choosing where to run a house show tour is a complicated calculation which involves previous tour attendances as well as proximity to other locations (with an emphasis on being able to drive to the next show).

Focusing on criteria of “recent House shows”, we’ve narrowed our dataset to 68 North American House Shows:

  •  Sat. 3/1/2014: Cedar Rapids, IA – US Cellular Center (capacity: 7,198, 70%)
  • Sun. 3/2/2014: Bloomington, IL – US Cellular Coliseum (capacity: 6,965, 48%)
  • Sun. 3/9/2014: Jackson ,TN – Oman Arena (capacity: 5,628, 58%)
  • Sun. 3/9/2014: Jonesboro, AR – ASU Convocation Center (capacity: 11,469, 61%)
  • Fri. 3/14/2014: Waco,TX – Extraco Events Center (capacity: 6,323, 81%)
  • Fri. 3/21/2014: Bridgeport, CT – Webster Bank Arena (capacity: 7,631, 100%)
  • Sun. 3/23/2014: Binghamton, NY – The Floyd L Maines Veterans Memorial Arena (capacity: 5,600, 69%)
  • Sat. 3/29/2014: Charlottesville, VA – John Paul Jones Arena (capacity: 12,319, 34%)
  • Sat. 4/12/2014: Hildalgo, TX – State Farm Arena (capacity: 6,545, 100%)
  • Sat. 4/26/2014: Louisville, KY – YUM Center (capacity: 16,914, 33%)
  • Sat. 4/26/2014: Rockford, IL – BMO Harris Bank Center (capacity: 6,129, 51%)
  • Sun. 4/27/2014: Evansville, IN – Ford Center (capacity: 10,643, 57%)
  • Sun. 4/27/2014: Peoria, IL – Peoria Civic Center (capacity: 10,942, 37%)
  • Fri. 5/2/2014: Rochester, NY – Blue Cross Arena (capacity: 10,700, 67%)
  • Sat. 5/3/2014: Syracuse, NY – War Memorial Arena (capacity: 6,484, 78%)
  • Sat. 5/31/2014: Springfield, IL – Prairie Capital.Convention Center (capacity: 8,324, 47%)
  • Fri. 6/6/2014: Detroit, MI – Joe Louis Arena (capacity: 14,494, 38%)
  • Sun. 6/15/2014: Erie, PA – Erie Insurance Arena (capacity: 6,532, 52%)
  • Sun. 6/22/2014: Philadelphia, PA – Wells Fargo Center (capacity: 15,913, 59%)
  • Sat. 6/28/2014: Bangor, ME – Cross Insurance Center (capacity: 5,988, 50%)
  • Sun. 7/6/2014: Toronto, ON – Air Canada Centre (capacity: 13,743, 51%)
  • Sat. 7/12/2014: New York, NY – MSG (capacity: 12,709, 68%)
  • Sun. 7/13/2014: Wildwood, NJ – Wildwood Convention Center (capacity: 7,000, 72%)
  • Sat. 7/19/2014: Tallahassee, FL – Donald L Tucker Civic Center (capacity: 8,175-11,624, 41%)
  • Sat. 7/26/2014: Edmonton, AB – Rexall Place (capacity: 13,310, 40%)
  • Sun. 7/27/2014: Calgary, AB – Scotiabank Saddledome (capacity: 12,004, 24%)
  • Sun. 8/3/2014: San Antonio, TX – AT&T Center (capacity: 13,639, 52%)
  • Sat. 8/16/2014: San Jose, CA – SAP Center (capacity: 12,600, 54-56%)
  • Fri. 8/22/2014: Stockton, CA – Stockton Arena (capacity: 8,535, 50%)
  • Sat. 8/23/2014: Sacramento, CA – Sleep Train Arena (capacity: 12,960, 47%)
  • Sat. 8/23/2014: San Diego, CA – Valley View Casino Center (capacity: 10,968, 41%)
  • Sun. 8/24/2014: Bakersfield, CA – Rabobank Arena (capacity: 8,745, 48%)
  • Sun. 8/24/2014: Fresno, CA – Save Mart Center (capacity: 10,689-11,168, 51%)
  • Sat. 9/6/2014: Hampton, VA – Hampton Coliseum (capacity: 7,978, 53%)
  • Sun. 9/7/2014: Salisbury, MD – WYCC (capacity: 5,775, 54%)
  • Fri. 9/12/2014: White Plains, NY – Westchester County Center (capacity: 4,375, 80%)
  • Sat. 9/13/2014: Jackson, MS – Mississippi St. Fair Coliseum (capacity: 9,261, 87%)
  • Sun. 9/14/2014: Boston, MA – TD Garden (capacity: 13,837, 43%-45%)
  • Sat. 9/20/2014: Huntsville, AL – Von Braun Civic Center (capacity: 6,969, 48%)
  • Fri. 10/3/2014: Toronto, ON – Ricoh Coliseum (capacity: 8,004-9,537, 64%)
  • Sat. 10/4/2014: Providence, RI – Dunkin Donuts Center (capacity: 8,606-9,538, 76%)
  • Sun. 10/5/2014: Bridgeport, CT – Webster Bank Arena (capacity: 7,631, 71%)
  • Fri. 10/17/2014: Montreal, QC – Bell Centre (capacity: 16,591, 29%)
  • Sun. 11/2/2014: Syracuse, NY – War Memorial Arena (capacity: 6,484, 49%)
  • Fri. 12/5/2014: Columbia, SC – Colonial Life Arena (capacity: 11,874, 52%)
  • Sat. 12/6/2014: Augusta, GA – Augusta Civic Center (capacity: 7,300, 63%)
  • Sat. 12/6/2014: Fairfax, VA – Patriot Center (capacity: 8,088, 34%)
  • Sun. 12/7/2014: Charlotte, NC – Time Warner Cable Arena (capacity: 14,650, 28%)
  • Sat. 12/20/2014: Sioux Falls, SD – Denny Sanford Premier Center (capacity: 6,549, 43%)
  • Sat. 12/20/2014: Winnipeg, MB – MTS Centre (capacity: 11,002, 58%)
  • Fri. 12/26/2014: Chicago, IL – Allstate Arena (capacity: 13,177, 68%-70%)
  • Fri. 12/26/2014: New York, NY – MSG (capacity: 12,709, 100%)
  • Sat. 12/27/2014: Cincinnati, OH – US Bank Arena (capacity: 10,955, 56%)
  • Sun. 12/28/2014: Hershey, PA – Giant Center (capacity: 8,777, 74%)
  • Sun. 1/4/2015: Beaumont, TX – Ford Arena (capacity: 7,422, 45%)
  • Sun. 1/4/2015: Hildalgo, TX – State Farm Arena (capacity: 6,564, 91%)
  • Sat. 1/10/2015: Sacramento, CA – Sleep Train Arena (capacity: 13,315, 45%)
  • Fri. 1/16/2015: St. Louis, MO – Scottrade Center (capacity: 14,569, 33%)
  • Sun. 1/18/2015: Houston, TX – Toyota Center (capacity: 11,580, 67%)
  • Fri. 1/23/2015: Trenton, NJ – Sun National Bank Center (capacity: 7,719, 68%)
  • Sat. 1/31/2015: Edmonton, AB – Rexall Place (capacity: 13,310, 49%)
  • Sun. 2/1/2015: Calgary, AB – Scotiabank Saddledome (capacity: 12,006, 26%)
  • Sun. 2/1/2015: Loveland, CO – Budweiser Events Center (capacity: 5,628, 62%)
  • Fri. 2/6/2015: North Charleston, SC – North Charleston Coliseum (capacity: 8,496, 70%)
  • Sat. 2/7/2015: Jacksonville, FL – Veterans Memorial Arena (capacity: 10,610, 57%)
  • Sat. 2/7/2015: Salt Lake City, UT – Maverik Center (capacity: 8,503, 55%)
  • Sat. 2/14/2015: Tampa, FL – Amalle Arena (capacity: 14,223, 44%)
  • Sun. 2/15/2015: Fort Myers, FL – Germain Arena (capacity: 8,175, 71%)

The next step is looking at who was wrestling on these cards and isolate the top stars and headliners.

To begin, I chose this list of wrestlers:

  • Dean Ambrose: 26 shows (average %: 60.0%)
  • Daniel Bryan: 10 shows (average %: 54.7%)
  • John Cena: 35 shows (average %: 61.7%)
  • Cesaro: 39 shows (average %: 55.6%)
  • Kane: 29 shows (average %: 55%)
  • Luke Harper: 40 shows (average %: 58.3%)
  • Miz: 33 shows (average %: 53.3%)
  • Randy Orton: 18 shows (average %: 62.5%)
  • Roman Reigns: 38 shows (average %: 56.2%)
  • Seth Rollins: 35 shows (average %: 58.9%)
  • Rusev: 36 shows (average %: 52.9%)
  • Sheamus: 25 shows (average %: 60.1%)
  • Big Show: 32 shows (average %: 55.0%)
  • Bray Wyatt: 44 shows (average %: 58.9%)
  • Dolph Ziggler: 36 shows (average %: 58.7%)

These wrestlers were chosen by looking at which wrestlers were in the last few matches most often. (I originally had Ryback included in this list, but I dropped him because his p-value ended up being insignificant and I knew everyone was going to ask me “What about Bryan? What about Bryan?”. I was doing a standard data analysis toolkit regression in Excel which limits to 16 variables.)

First step? Create a matrix of the wrestlers and look at how filled the arenas were.

wwe_matrix_houseshows

Click to Expand

 

Check.

Next step: perform a regression between which wrestlers were on the card (1=booked, 0=Dennis Stamp’d) and see how that correlates to the percentage filled of the arena.

wwe_regression1

Reading other people’s messy stat regressions is as exciting as actually being at a Twins game. It’s dull.

The Results (First Pass)

Positive influence: John Cena, (+14%, low p-value), Bray Wyatt (+6%, high p-value), Roman Reigns (+5%, high p-value), Dolph Ziggler (+5%, high p-value), Big Show (+4%, high p-value), Randy Orton (+4%, high p-value)
Little influence: Sheamus (+2%, very high p-value), Dean Ambrose (0%, exceptionally high p-value), Usos (-1%, exceptionally high p-value), Kane (-2%, very high p-value), Seth Rollins (-3%, very high p-value)
Negative influence: Daniel Bryan (-4%, very high p-value), The Miz (-5%, high p-value), Cesaro (-5%, high p-value), Luke Harper (-8%, high p-value), Rusev (-9%, medium p-value)
Intercept: 55%

For this bout of lazy Wrestlenomics, think of the p-value as a measure of how much we can trust this result. Typically you’re looking for p-value that is below 0.05 or even 0.01. Here, we’re lucky if we have anyone that’s even below 0.10. Not a good sign.

Furthermore, look at that adjusted r-squared value: 0.01.  That’s really, really poor. Basically, we can take it to mean that there wasn’t a correlation between using all of these variables and predicting the outcome.

What do we do? First, let’s eliminate the wrestlers with the absolute highest p-values (Orton, Bryan, Rollins, Kane, Sheamus, Usos and Ambrose) and try again.

wwe_regression2

When I lived in West Africa, I got malaria. True story.

Now we’re getting a little closer. (Note that the adjusted r-squared has risen to 9.6%. Still not great, though.)

The Results (Second Pass)

Positive influence: John Cena (+16%, low p-value), Roman Reigns (+6%, medium p-value), Bray Wyatt (5%, medium p-value)
Little influence: Dolph Ziggler (3%, high p-value), Big Show (+2%, high p-value)
Negative influence: The Miz (-4%, high p-value), Cesaro (-6%, medium p-value), Rusev (-8%, low p-value), Luke Harper (9%, medium p-value)

Again, let’s eliminate the variables with the highest p-values (Big Show, Dolph Ziggler, The Miz, Bray Wyatt) and re-run our regression.

wwe_regression3

I think Night Court and Newsradio are my favorite sitcoms.

Here’s where we landed: adjusted r-squared of 11.5%.

The Results (Final Pass)

Intercept: 55.2%
Positive influence: John Cena (+17.2%, very low p-value), Roman Reigns (+6.3%, medium p-value)
Negative influence: Rusev (-7.5%, low p-value), Luke Harper (-6.9%, medium p-value), Cesaro (-5.3%, medium p-value)

What did we learn?

John Cena is good for House Shows. (I know, shocking!) If the average capacity of these arenas was 9,910, Cena would be worth about 1,700 more people which is a very respectable amount.

If you line up the ten weekends where there was two tours (and we have data on shows), the show with Cena on it averaged 64% while the show without Cena averaged 55%.

Some of the variation with people is due to residual effects of who they were wrestling. For instance, when Bray Wyatt was on the same show as Cena, the average percentage filled was 62% but when Bray was on a show not with Cena, the average dropped to 53%.

This approach looks the capacity of each wrestler to fill an arena. However, 1% doesn’t mean the same thing in every city. (For Salisbury, Maryland that’s 58 people while in Montreal that could be 166 people-almost triple.)

Do we get different results if we run the regression against the crowds instead of the percentage of capacity filled?

The Rochester Roadblock is technically neither from Rochester nor a Roadblock.

The Rochester Roadblock is technically neither from Rochester nor a Roadblock.

Not sure what changed, but now we have intercept of 3,909 (about 40%) with Bray Wyatt (+1,230), Randy Orton (+1,008), John Cena (+838) and Miz (+615) on one side of the equation and Daniel Bryan (753) and Sheamus (-648) on the other. If you keep whittling down the list, you end up with just the equation: drawing power = 4,404+Bray Wyatt (+1,065) + Randy Orton (+983).

Maybe it’s a sign that WWE needs these two two feud.

Chris Harrington is the junior Wrestlenomics intern for Voices of Wrestling. You can always find him on Twitter (@mookieghana) or email (chris.harrington@gmail.com).