Is it possible to use Wrestlenomics to isolate the “influence” that each wrestler has on the overall goodness of a wrestling match? That’s the question I’m trying to tackle. Here’s how I approached the subject.
The data set used included Wrestling Observer Newsletter (WON) Star Ratings covering 3,651 matches
- WWF/WWE/NXT PPV and SNME/Main Event TV & Network specials (1985-2015): 2,428 matches
- NWA/WCW and Clash of Champions TV specials (1986-2001): 1,223 matches
Isolated 2,413 singles matches with WON (1,617 WWF/WWE/NXT, 796 NWA/WCW) where competitor A wrestled competitor B.
Objective is to create an Wrestlenomic measure that estimates the “influence” that each wrestler has on a match. How does this model work? It may be easiest to give a single example and follow it through.
Let’s look at the Mark Henry versus Kurt Angle feud:
- Royal Rumble 2006: Kurt Angle defeated Mark Henry (w/Daivari) (7:00) 3/4*
- WWE Judgment Day 2006: Mark Henry defeats Kurt Angle by Count Out (9:11) **1/2
First, let’s see each man’s overall average WON rating:
- Among singles matches Mark Henry averaged 1.167 WON stars over 27 singles matches.
- Among singles matches Kurt Angle averaged 3.132 WON stars over 51 singles matches.
If you exclude these two Henry versus Angle matches, Mark Henry average WON stars drops to 1.130 WON stars over 25 singles matches (a drop of 0.037 WON stars) and Kurt Angle average WON stars rises to 3.194 WON stars over 49 singles matches (an increase of 0.062 WON stars).
Thus, when Mark Henry wrestled Kurt Angle, Angle’s overall average WON stars went down and Henry’s overall WON stars went up. We’re going to allocate the “influence” of Kurt Angle over the two match feud as +0.037 and the “influence” of Mark Henry over the feud as -0.062.
(You’ll notice that the changes aren’t equal. How many overall singles matches are in the data and many matches were in that particular singles feud for each competitor is going to skew how large an impact that difference calculations will be.)
We continue this exercise for all of the singles matches in the data set and we add the “influence” factors for all of the wrestlers.
Influence Calculation (Version 1)
If you were to put the competitors into four buckets they would be:
STRONG POSITIVE INFLUENCE (more than 20 singles matches)
Bret Hart (0.11), Brian Pillman (0.10), Billy Kidman (0.09), Chris Benoit (0.08), Christian (0.07), Ricky Steamboat (0.07), Ric Flair (0.06), Shawn Michaels (0.05), Chris Jericho (0.05), Randy Savage (0.05), Dean Malenko (0.04), Eddie Guerrero (0.04), Lex Luger (0.04)
POSITIVE INFLUENCE (more than 20 singles matches)
John Cena (0.03), Diamond Dallas Page (0.03), Matt Hardy (0.02), Vader (0.02), Rey Mysterio (0.02), Edge (0.01), Hulk Hogan (0.01), The Rock (0.01), Sheamus (0.01), Jeff Hardy (0.01), Marc Mero (0.01)
LITTLE INFLUENCE (more than 20 singles matches)
Randy Orton (0.00), Dustin Rhodes (0.00), Triple H (0.00), Alberto Del Rio (0.00), CM Punk (0.00), Chavo Guerrero Jr (0.00), Bam Bam Bigelow (0.00), The Undertaker (0.00), The Miz (0.00), Rob Van Dam (0.00), Scott Steiner (0.00), Curt Hennig (-0.00), Sting (-0.00), Brock Lesnar (-0.00), Goldberg (-0.00), Booker T (-0.00), Batista (-0.01), Bradshaw (-0.01), William Regal (-0.01)
NEGATIVE INFLUENCE (more than 20 singles matches)
Kevin Nash (-0.02), Kane (-0.02), Paul Wight (-0.03), Scott Hall (-0.04), Rick Rude (-0.04), Sid Vicious (-0.04), Mark Henry (-0.04)
Were the initial results surprising?
I expected that this analysis of WON single match star ratings would highly rate people such as Bret Hart, Ric Flair, Ricky Steamboat, Shawn Michaels, Chris Jericho and Chris Benoit. These are all wrestlers who have been praised lavishly by Meltzer for their working ability and how they were capable of extracting great matches from limited opponents. Likewise, I was not surprised to see Sid Vicious, Big Show or Kane at the bottom end of this “influence” list.
I am very surprised to see some other “worker” candidates such as Kurt Angle, Triple H, William Regal, CM Punk, Curt Hennig and Edge falling to the middle of the distribution. Likewise, the placement of Lex Luger and Jeff Jarrett surprised me for how high they finished. (We’ll dig into that Lex example a little further down.)
Still, let’s also put this in perspective – consider that star ratings typically work in quarter (0.25) increments. There’s a small handful of people that would appear to be able to actually push any star ratings up or down based on their average “influence”. We might see a slight nudge up for a Brian Pillman or Christian singles match but not much else.
Where’s the rest?
This 57 person list feels bit light, doesn’t it? That’s because it was filtered on those with more than twenty one-on-one WON rated singles matches from WWE/WCW PPVs, TV specials and WWE Network events. There’s a host of other wrestlers (over 500) who were excluded with this filter. Among the 78 wrestlers with ten to twenty matches, you’d add in:
STRONG POSITIVE INFLUENCE (10-20 matches)
Terry Funk (0.39), Great Muta (0.18), Kevin Sullivan (0.17), Mike Awesome (0.11), Ted DiBiase Sr. (0.11), Barry Windham (0.09), Juventud Guerrera (0.09), Beth Phoenix (0.07), Cesaro (0.07), Tom Zenk (0.06), Tito Santana (0.06), Melina (0.05), Davey Boy Smith (0.05), Ultimo Dragon (0.05), Daniel Bryan (0.04), Michelle McCool (0.04)
POSITIVE INFLUENCE (10-20 matches)
Paul Orndorff (0.03), Shane McMahon (0.03), Raven (0.03), Kevin Owens (0.03), Terry Taylor (0.03), Seth Rollins (0.03), Mike Rotunda (0.02), Vince McMahon (0.02), Kofi Kingston (0.02), Trish Stratus (0.01), John Morrison (0.01), Fit Finlay (0.01), Arn Anderson (0.01)
LITTLE INFLUENCE (10-20 matches)
Bob Holly (0.00), Shelton Benjamin (0.00), Bobby Eaton (0.00), Rusev (0.00), Ultimate Warrior (-0.00), Buff Bagwell (-0.00), Madusa (-0.00), Montel Vontavious Porter (-0.00), Al Snow (-0.00), Bray Wyatt (-0.00), Val Venis (-0.00), Perry Saturn (-0.00), Greg Valentine (-0.00), AJ Lee (-0.01), Jerry Lawler (-0.01), Rick Steiner (-0.01), Ken Shamrock (-0.01), Mr Kennedy (-0.01), John Tenta (-0.01), Brutus Beefcake (-0.01), Owen Hart (-0.02), Brad Armstrong (-0.02), Umaga (-0.02), Ryback (-0.02)
NEGATIVE INFLUENCE (10-20 matches)
Alex Wright (-0.03), Paige (-0.03), Dean Ambrose (-0.03), Cody Rhodes (-0.03), Jack Swagger (-0.04), Konnan (-0.04), Mickie James (-0.04), Yokozuna (-0.04), Jake Roberts (-0.04), Rikishi (-0.05), Meng (-0.05), Ron Killings (-0.05), Honky Tonk Man (-0.06), Wade Barrett (-0.06), Steve McMichael (-0.07), Jim Duggan (-0.08), Ron Simmons (-0.09), Viscera (-0.09), Roddy Piper (-0.09), Great Khali (-0.10), Carlito (-0.12), Ernest Miller (-0.12), Disco Inferno (-0.12), Dusty Rhodes (-0.15), Vampiro (-0.15)
Again, this list is a mix of expected (Danielson, Cesaro, Windham had positive influence, Khali, Mongo and Viscera had negative influence) and unexpected (Melina and Mike Awesome so “high”, Brad Armstrong and Dean Ambrose so “low”). It’s always important to take things with a grain of salt. The lower number of matches tends to lead to overstating influence.
The 5-9 match list has your strong positive influences (Ricky Morton, Adrian Adonis, Damien Sandow, Sasha Banks, Finn Balor, Victoria, Charlotte, Sami Zayn, Tajiri, Torrie Wilson, Kanyon, Molly Holly, Neville, Rhyno, Scorpio, Natalya, Harley Race) and strong negative influences (Don Muraco, Eric Bischoff, Bryan Clarke, Billy Gunn, Candice Michelle, Savio Vega, Big E, Layla, David Flair, Matt Borne, One Man Gang, Stevie Ray, Ahmed Johnson, Eve Torres, Van Hammer, The Wall, Prince Iaukea, Sable, Dennis Knight, D-Lo Brown, Ivory, Dave Sullivan, Dino Bravo, Fandango, Butch Reed, Junkyard Dog).
There’s also the issue of negative match ratings and the dreaded DUD. (Royal Rumble 1993 and 2005 got a DUD rating as did several twenty minute matches including Tonga/Muraco at the 1986’s “The Big Event”, Dr. Death/Koloff at GAB 92 and that Uncensored 1997 match with Piper/nWo/WCW teams.) What happens if we just exclude those matches from the data set?
The eight people who change the most when you exclude the sub-zero and DUDs were Tito Santana, Mickie James, Roddy Piper, Davey Boy Smith, Tom Zenk, Mike Rotunda, Great Muta and Greg Valentine. The majority of these people had less than twenty matches, so small inclusions or exclusions is going to explain a lot of difference.
On that topic, let’s take little closer look at Lex’s +0.04 influence.
The five matches that had provided the largest “positive influence” examples for Luger (0.04 average influence in the original calculation) were:
- NWA WrestleWar 1989: Luger vs. Hayes *** (+0.75 influence)
- NWA Clash Of The Champions #8: Lex Luger vs. Tommy Rich **** (+0.35 influence)
- NWA Halloween Havoc 1990: Hansen vs. Luger **1/2 (+0.31 influence)
- NWA Starrcade 1990: Hansen vs. Luger ***1/4 (+0.25 influence)
- WCW Starrcade 1995: Luger vs. Chono *1/4 (+0.25 influence)
What do these five matches have in common? Luger’s opponents have very few matches in the data set. Hayes only had two (i.e. only one non-Luger match, a singles match against Russian Assassin #1 at Chi-Town Rumble), Rich had nine, Chono had four and Hansen had four. It’s easier to get a high “influence” rating when the baseline you’re being compared against is very small. If you exclude those five matches, suddenly Lex’s influence drops to +0.01. When you exclude all of the people who had less than ten matches in the data set, Lex Luger’s influence drops to -0.03.
Influence Calcuation (Version 2)
Let’s refine our criteria once again. Now we’re looking at:
- Influence Calculations on singles (A vs B) matches
- Excluding matches with DUD or negative star ratings
- Excluding matches involving anyone who have less than ten matches in the dataset
Remaining current singles data set: 1,151 matches.
But what factor does match length play on Star Rating?
It may be possible to try and compensate for match length. Consider this:
|Less than 4:06||240||0.32|
|More than 19:30||241||3.56|
This can be a bit of a self-fulfilling prophesy. Top wrestler with proven track records may be given longer matches while weaker wrestlers may be intentionally given short matches.
A logarithmic regression on this data suggests that on average a 4:47 match will be rated * (one star), an 8:33 match will be rated ** (two stars), a 15:18 will be rated *** (three stars) and 27:23 match will be rated **** (four stars). In this model, it’s a not a linear time difference between star ratings (it takes almost four minutes to move from one to two stars but almost seven minutes to move from two to three stars and more than twelve minutes to move from three to four stars).
We can try to incorporate this information into the model.
Let’s go back to those Kurt Angle versus Mark Henry matches.
At Royal Rumble 2006, the bout was only seven minutes and earned a 0.75 star rating. The expected star rating (average star rating for a match of this length) would be about 1.66 stars, so the match underperformed by 0.91 stars.
At Judgment Day 2006, Henry’s count out victory had a 2.5 star rating in 9:11. The expected star rating for a match of this length would be 2.12 stars, so the match overperformed by 0.38 stars.
As noted before, time allocation isn’t random nor is it evenly distributed. Big Boss Man’s matches averaged 8:26 in his singles matches with the longest one clocking in at 14:15 (an encounter with the Barbarian Royal Rumble 1991 that earned 2.75 stars, just shy of the 2.88 stars that would have been predicted). Conversely, Sheamus’ singles matches averaged 14:19 with the longest one clocking in at nearly 24 minutes (retaining his title at TLC 2015 against Roman Reigns which earned 4.0 stars, almost a full quarter-star above the 3.77 Star-Time prediction.)
The most time-effective performers? Steamboat, Pillman, Malenko, Rey, Benoit, Christian, Ziggler, Waltman, Guerrero, Foley, Angle, Jericho, Vader, Chavo, Edge, Matt Hardy and Savage.
The least? Nash, JBL, Hogan, Mark Henry, Kane, Undertaker, Goldust, Luger, Big Show, HHH, Booker, Regal and Sting.
What’s very interesting is someone like Bret Hart, who had very high influence ratings, has virtually no difference between the time-predicted star rating and actual star rating. The same can be said for John Cena and Diamond Dallas Page. Similarly, The Rock, Ray Traylor, Scott Hall, Brock Lesnar, The Miz, Goldberg and Curt Hennig typically deliver on par with the expected star rating despite having a no or even negative influence calculation.
Now, we can repeat the “influence” methodology, but instead of using Star Ratings, we’ll use the delta between Actual Star Rating and the Time-Predicted Star Rating. Essentially, we’re creating the expectation that if you’re given more time, you should have a better match. This will be done on the positive, non-DUD star ratings and only among people with ten or more matches in the dataset.
POSITIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (20+ matches)
Randy Savage (+0.028), Steve Austin (+0.025), Chris Benoit (+0.022), Diamond Dallas Page (+0.020), Vader (+0.020), Christian (+0.018), Bret Hart (+0.016), Rey Mysterio (+0.016)
SLIGHT POSITIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (20+ matches)
Goldberg (0.014) , Sheamus (0.014) , Eddie Guerrero (0.013) , Mick Foley (0.013) , Shawn Michaels (0.012) , Dolph Ziggler (0.01) , The Rock (0.009) , Kurt Angle (0.008) , Jeff Jarrett (0.008) , Edge (0.007)
FLAT IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (20+ matches)
Chris Jericho (0.004) , Marc Mero (0.004) , Alberto Del Rio (0.003) , Brock Lesnar (0.002) , Ric Flair (0.002) , Sting (0.001) , CM Punk (0.000) , The Miz (0.000) , John Cena (-0.002) , Rob Van Dam (-0.003) , Jeff Hardy (-0.004)
SLIGHT NEGATIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (20+ matches)
Triple H (-0.007) , Randy Orton (-0.008) , Batista (-0.01) , Big Show / Giant (-0.011) , Mark Henry (-0.012) , Sid Vicious (-0.013) , Booker T (-0.014)
NEGATIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (20+ matches)
Bradshaw (-0.015) , Undertaker (-0.016) , Kane (-0.027) , William Regal (-0.029) , Kevin Nash (-0.029) , Hulk Hogan (-0.031) , Lex Luger (-0.034) , Dustin Rhodes / Goldust (-0.035)
What’s remarkable about this is how small we’re talking. Randy Savage would make a match about 0.03 stars better than the time would have expected. Goldust would have made it about 0.03 star worse. That’s just a rounding error when you realize we’re only talking about quarter-star increments.
For posterity, the other wrestlers who have less than twenty matches left in this data set:
POSITIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (<20 matches)
Owen Hart (0.042), Rick Rude (0.033), Daniel Bryan (0.023), Seth Rollins (0.022) , Ricky Steamboat (0.021), Kofi Kingston (0.017), Matt Hardy (0.017), Rusev (0.015) , Raven (0.015) , Sean Waltman (0.015)
SLIGHT POSITIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (<20 matches)
Shelton Benjamin (0.014), Ken Shamrock (0.01), Roddy Piper (0.009), Cody Rhodes (0.009), Davey Boy Smith (0.009), Brian Pillman (0.008), John Morrison (0.008), Yokozuna (0.006), Dean Ambrose (0.005), Scott Steiner (0.005)
FLAT IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (<20 matches)
Fit Finlay (0.004), Umaga (0.002), Jack Swagger (0.002), Wade Barrett (0.001), Perry Saturn (0.001) , Dean Malenko (0.001), Ryback (0.000)
SLIGHT NEGATIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (<20 matches)
Montel Vontavious Porter (-0.003) , Ultimate Warrior (-0.003) , Chavo Guerrero Jr (-0.004) , Bray Wyatt (-0.005)
NEGATIVE IMPACT ON MATCHES AFTER FACTORING IN MATCH LENGTH (<20 matches)
Curt Hennig (-0.01) , Buff Bagwell (-0.011) , Scott Hall (-0.013) , Ray Traylor (-0.015) , Rick Steiner (-0.022) , Great Khali (-0.028) , Meng (-0.035) , Jim Duggan (-0.058)
Again, most of the people I’d expect rise to the top and most of the people I’d expect sink to the bottom.
Nothing is guaranteed. Ted DiBiase and Jim Duggan can hang a Coal Miner’s Glove at the top of Steel Cage and wrestle in Tuxedos in a Loser Leaves Town Match and blow the roof off the joint. Triple H can go 38 minutes against Shawn Michaels and end up with a two star match.
Other commentary on shortcomings on this analysis:
- This is based on the ratings data from a single source (Wrestling Observer Newsletter) and reflects the viewpoint, biases, likes, dislikes and workrate lens of a single author (Dave Meltzer).
- Other ratings from major professional wrestling companies—namely ECW, TNA, SMW, NJPW, AJPW, NOAH, DG, ROH, AAA and CMLL—were not included in this analysis. For instance, there would be a lot of data on the newer crop WWE talent (Finn Balor, Kevin Owens, Samoa Joe, Neville, Sami Zayn) if ROH, TNA, DG and NJPW results were included.
- This is only incorporating PPV, WWE Network and certain TV specials. While this provides some sense of balance (more of the “major” matches), it leaves out a host of rated television matches. As the professional wrestling business evolved during the Monday Night Wars, more and more of the big-time matches were presented during the weekly RAW & Nitro episodes. Unfortunately, providing star ratings for television matches in the WON has been a very uneven phenomenon. Often, if a match is quite spectacular, it will receive a rating while most of the hum-drum matches do not. These sort of positive selective ratings run the risk of inflating overall ratings for top performers.
- This does not include any multi-man matches including tag team matches, rumbles, battle royals, gauntlet, or multi-man singles matches (such as Elimination Chamber or Money in the Bank).
- This only covers the time-frame 1985 through 2015 with less than 7% of the matches from the 1980’s.
- This analysis is not incorporating the age, experience and physical condition of the competitors. Some of my earlier analysis of star ratings had suggested that a pro-wrestler typically has their best matches when they are in the mid-30s. If you’re looking at a wrestler like Dusty Rhodes, who was already forty in 1985, this dataset is likely looking at their performance while they are on the decline in ability.
Special thanks for Alex Sarti (@alexwkpartyhard) for helping compile thirty years of WON Ratings.