3/01/2017

What is the best time of the year for quality WWE wrestling?

This is a short one. The question is simple: what level of wrestling quality can you expect from the WWE in different months? We look at Wrestling Observer Newsletter star ratings for WWE PPV matches from April 2002 until now (April 2002 saw the WWF become the WWE and that sort of marks the beginning of the post-attitude era, but really it just functions as an arbitrary cut-off point for me to do the analysis).

The following graph shows the average WON star rating for a WWE PPV match between 2002 and now which is around **1/2 and is thus right in the middle of the *****-rating continuum. (BTW: star ratings are not normally distributed around the mean but the distribution is skewed in that there are substantially more bad [* and lower] than great [>= ****] matches - sadly 😒; what pushes the mean towards the center are a large share of ***-to***1/2 matches).
Anyway, the blue-ish line shows the average rating for matches which have taken place at a WWE PPV in the respective month:

Takeaways:

  • Wrestlemania rocks: March and April usually mean good times for fans of good wrestling - that's not very surprising as these months regularly include Wrestlemania which tends (see my post on Wrestlemania quality over time below) to deliver quality wrestling!
  • Summerslam is pretty good, too: Also not surprising as this is widely considered WWE's number two event (solidly behind Wrestlemania, though) and usually takes place in August
  • Survivor Series blows: Well, I hadn't expected this but thinking about it, I can't really remember a great Survivor Series PPV... November is the second worst month for WWE PPVs
  • Christmas and Wrestling = no match: WWE's final PPV of the year tends to be a real stinker, scoring way below the average!
    • Armageddon ... yikes!
  • So - as a general rule - if you need a break from the WWE (which really helps from time to time), you might as well take it between September and January. This is especially handy if you're an NFL fan!
    • Great months: March, April, August
    • Average months: January, February, May, June, July, September
    • Awful months: October, November, December


2/27/2017

The development of NXT Takeover

Let's take a quick first look at Triple H's (and probably everyone else's except Vince) favourite pet project, NXT!

Well-deserved credit: the data for this analysis come from profightdb.com!

I am a big fan of NXT and especially of the recently established NXT Takeover PPV-like big events of WWE's developmental promotion. And that's all the reason I need to look at some NXT data and show you some hopefully interesting stuff. :)

The following graph is based on Wrestling Observer Newsletter Star Ratings of NXT matches and shows for each Takeover event the rating for the best (yellow line), worst (red-ish line), and average (blue-ish line) match of the show. The events are obviously sorted by date - with the most recent Takeover: San Antonio at the right side of the graph and the first Takeover event ArRival at the left side of the graph.

Takeaways:
  • NXT is constantly getting better: NXT started strong with NXT ArRival (average match rating of **1/2 with a ****1/4 showstealer by Sami Zayn and Antonio Cesaro) but has even gotten better and better since when it comes to the overall quality of the shows (based on the average match rating)!
  • NXT TakeOver: Toronto is the best NXT TakeOver event to date: no match was worse than *** and the 2 out of 3 falls tag title match between The Revival and #DIY was the match of the night with an impressive ****1/2 rating
  • NXT has always delivered at least one really strong match: This is really ridiculous and reminds me of 2004-2006 Ring of Honor. Every single Takeover Event has had at least one match cracking the ****-ceiling!
    • In this regard, NXT Takeover have stayed on the same (very high) level over time
  • NXT is becoming an even more well-rounded product: What you notice most from the graph is that the three lines move closer to each other over time - and thankfully this is not because the best match on a Takeover card is getting worse but because other matches are getting better!
    • For the last eight Takeover events, the rating average has been above ***!
    • Even the worst matches are now typically no worse than **!

That's it for this post. More NXT to come!

2/25/2017

Royal Rumble: STUD or DUD? The top and flop performers.

Again, let's look at every data-focused person's favourite WWE event, the Royal Rumble!

This post is focused on finding Royal Rumble studs ... and duds. Only wrestlers who have participated in at least five Rumble matches are considered because we really want to focus on Rumble regulars here.

Speaking of Rumble regulars, Kane has entered 16 Rumble matches. Sixteen! He has also eliminated a record of 40 other wrestlers. But to no avail - Kane has never won the Royal Rumble. 👿

People who have won the Royal Rumble more than once (instant STUDs, I guess): Steve Austin (3), Triple H (2), Randy Orton (2), Batista (2), Hulk Hogan (2), Shawn Michaels (2), John Cena (2) ... holy cow, Evolution is likely the only wrestling stable in history where every member had already won or would go on to win the Royal Rumble.

(Bad) Luck of the draw

  • Mr. Bad Luck (DUD): CM Punk has entered 6 Rumble matches and won none of them. This might have been due to his bad luck with the draw. On average, CM Punk entered the Rumble match at position 8. Randomness would have you enter at position 15 on average. Poor Punk!
  • Mr. Luck (STUD): Batista has participated in 5 Rumble matches and won two of them. No wonder as his average entry number has been number 24! He was booed out of the arena in 2014 anyway.


How lang can you survive?

  • Marathon men (STUD): While Chris Jericho has spent an impressive 4 hours and 51 minutes overall in Royal Rumble matches, Steve Austin - having entered five Rumble matches, and winning an impressive three - has averaged 30 minutes in Rumble matches.
  • Not very long (DUD): Farooq has entered 5 Rumble matches and spent a total of 12 minutes in the ring. Ouch! However, Farooq still averages about two and a half minutes per Rumble match. That's half a minute more than The Great Khali!


How many opponents can you eliminate?

  • Not that many, actually (DUD): While he did participate in 6 Rumble matches, Jack Swagger has yet to eliminate somebody!
  • Beast (STUD)Steve Austin eliminates an incredible 7 opponents per Rumble! While Kane does hold the record for alltime eliminations, it took him 16 Rumbles to get there. Kane's average is actually 2.5 eliminations (same as Randy Orton's, by the way).


Efficiency: How many opponents can you eliminate per ten minutes?

  • Mr. Efficiency (STUD): Oh SWERVE! The Great Khali eliminates 6 people per ten minutes. Only problem is that he never lasted more than 7 minutes in a Rumble match! The true Mr. Efficiency is Big Show who has spent a respectable 72 minutes in Rumble matches and eliminates 4 people per ten minutes (Steve Austin, by comparison, eliminates 2 opponents per ten minutes)!
  • Forget about it (DUD): Well, Jack Swagger has never eliminated anybody so he 'wins' this by default. However, the 'runner-up' is The Miz who eliminates an opponent every 60 minutes! Wow, that is brutal. Big Show eliminates 24 people for Miz's 1. Looking at these guys, that does not sound unreasonable, actually. 



Honorable(?) mention: Roman Reigns has not yet participated in 5 Royal Rumble matches but he has to be mentioned here. Man, yeah, he is THE GUY, we get it. We really do! The pace at which Reigns is going should see him breaking various Rumble records! He has eliminated 26 people in 4 Rumbles (putting him slightly behind Austin in terms of average eliminations) already ...

2/22/2017

Wrestlemania: relationship between match length and match quality

We all - at least in our minds - have chanted "FIGHT FOREVER!" now and again. And Bret Hart and Shawn Michaels pretty much did at Wrestlemania 12 in 1996! But how are match length and match quality actually related statistically?

Obviously, the WWE is not completely crazy and usually puts those wrestlers in long matches who are fairly good in the ring. And, well, Triple H always gets long matches - sometimes he's fairly good and then sometimes he's not so good (especially vs. Randy Orton). So let's look at all Wrestlemania matches and see how match length and match quality (star rating based on Keith and Meltzer) are related:

Note: Hart/Michaels from Wrestlemania 12 (61 minutes) is not shown in the graphs because it really messes with their dimensions being such a strong outlier. For completeness sake: match got an average 4.5* rating.




(Somewhat premature) Takeaways:
  • Clear general trend: Longer matches tend to be better matches as wrestlers get more time to tell a great story and as there is a serious selection bias by which good wrestlers are usually (hello, Hulk Hogan and JBL) given more in-ring time than bad wrestlers!
    • The correlation between both measures is a very solid +.7 (https://en.wikipedia.org/wiki/Correlation_coefficient)
  • A DUD is a DUD: What is apparent from the many dots across the bottom of the graph is that you can have 10 minutes or longer and still produce a sh*t match!
HOWEVER! This trend (more time = more quality) is only strong for matches going 15 minutes and less. The following graph shows only those matches going longer and what you see is not really all that clear. There is still a small positive correlation (+.25) but the general message seems to be that once you get to that area where you have a fairly long (15 minutes +) match, the length of the match is not really saying all that much with regards to match quality anymore!



There have been 40 20+ minutes in Wrestlemania history. 23 of those matches included at least one of the group of Triple H, Undertaker and Shawn Michaels!
  • While Taker and HHH of course had various classic matches (including against each other in a Hell in a Cell match), both occasionally stunk up the join with long and boring matches
    • HHH vs. Batista, Randy Orton, Brock Lesnar
    • Taker vs.  Shane McMahon, Brock Lesnar, Sycho Sid ... Brock Lesnar has thus had long dull Wrestlemania matches both against Triple H and the Undertaker - that's quite an accomplishment in its own right!
  • The only one who seems to be untouchable in long Wrestlemania matches is ... Mr. WRESTLEMANIA HBK Shawn Michaels! Even his worst 20+ minutes match against Ric Flair at Wrestlemania 24 rates at a good 3.375 stars. And the atmosphere throughout the match was so great that you could even argue that the star rating is a bit low to begin with.

Finally, don't forget that this post has only considered Wrestlemania matches. We'll later revisit this question and look at data from other events.

2/21/2017

Rumble draws and success chances

There is no WWE PPV more suited for intensive data analysis than the Royal Rumble!

The question at hand is how a wrestler's draw (number of entry) affects his (occasionally, her) chances of winning the Rumble match or at least lasting for a long time.

The WWE never tires to remind us that number 27 is the magic number for Rumble success - 4 Rumble winners have entered the Royal Rumble match as the 27th entrant (Big John Studd in 1989, Yokozuna in 1993, Bret Hart in 1994 and Stone Cold Steve Austin in 2001). 2001? Ok, number 27 does seem to be experiencing a bit of a dry spell these past 16 years ...
Furthermore, while technically each has produced 'only' two Rumble winners, numbers 1 and 2 really can be counted as one because both wrestlers 1 and 2 start the Rumble match at the same time (duh ...). Hence, could it be that one should hope to be among the first two wrestlers in a Rumble match?

Let's look at how long a wrestler can be expected to last based on his number of entry. The following graph depicts at which position each draw (from 1 to 30) can be expected (on average) to be eliminated from the Rumble match. E.g. draw number 5 is expected to last until the sixth elimination.


Takeaways:

  • You see the wrestling magic right there at the top of the graph: Numbers 1 and 2, on average, last until the tenth/eleventh elimination! So in the WWE you would want to trade every number between 3 and 10 for a starting spot. But you should prefer number 11 and onwards to the starting gig.
  • All other draws (3 and onwards) follow a very linear pattern where the higher you get the better you're off. It is funny to note, though, that there is a noticeable dent for entry number 27 casting more doubt on the notion of it being the magic Rumble number. Based on these results you should prefer numbers 25 and 26 over number 27.
  • If you can, go with number 30! Number 30, on average, means that you'll last the longest in the Rumble. Three people have won the Rumble from number 30: Undertaker in 2007, John Cena in 2008 and Triple H in 2016. Take that, 27!

2/20/2017

Wrestlemania: quality of the in-ring product

Still focusing on Wrestlemania!

After having seen that Wrestlemania matches have tended to get much more actual in-ring time over the years, let's now take a look at an arguably more interesting question: How has the quality of the in-ring product developed over time?

I have collected star ratings from two of the most-renowned WWE reviewers - Scott Keith (see http://blogofdoom.com/) and Dave Meltzer (Pro Wrestling Observer) - for all 32 Wrestlemanias and have computed the average rating for every single match.

Fun Fact: The average rating for a Wrestlemania match is two stars (2.04 to be precise). Ok, in isolation this isn't really saying a lot ... we'll get to a comparison of this average to the average of other events later. ;-)

The graph below shows two different values for every Wrestlemania. First, the average match rating is shown (darker blue) and second, the average in-ring minute rating is shown (light blue). The latter measure takes into account whether a good or a bad match took a lot of time. Hence, this latter measure will differentiate between a Wrestlemania which had two matches combining for 40 minutes of bad wrestling and a Wrestlemania which had two matches of bad wrestling but which combined only for 10 minutes.



Key takeaways:

  • Clear trend of rising quality between Wrestlemania I (1985) and Wrestlemania X-7 (2001): No matter which measure one focuses on, the in-ring quality has improved strongly during that time frame
    • Notable exception: Wrestlemania III (1987) was way ahead of it's time in terms of in-ring product quality (Thank you, Ricky Steamboat and Randy Savage!)
  • Best Wrestlemania: The usual suspect is the widely-loved Wrestlemania X-7 (2001). However, according to this analysis of the in-ring product it is beaten by another! The best minute-by-minute in-ring product has been produced by Wrestlemania 28 (2012)! That Wrestlemania had Triple H/Undertaker, Jericho/Punk and Rock/Cena and very little crap besides.
    • When it comes to average match quality, Wrestlemania 25 (2009) is the best Wrestlemania. I believe the twenty minute snorefest of a main event between Triple H and Randy Orton have really dragged the memory of that Wrestlemania down, because the numbers don't lie (and Taker/HBK was brillant stuff)!
  • Last year's Wrestlemania was pretty bland: When I first saw the results for the average match measure with last year's Wrestlemania 32 finishing in the top 3 all-time I was shocked. No way could that be a top 3 Wrestlemania - in no category! This feeling is corroborated by the more nuanced per-minute results: the average minute of Wrestlemania 32 was pretty mediocre. According to the average minute measure, Wrestlemania 32 has been the worst Wrestlemania since Wrestlemania X-8 (2002) which was truly awful. Yeah, that is certainly more in line with how I remember it!
    • There is a price to pay for 60 minutes of Shane McMahon trying to murder himself in order to make completely broken Undertaker look somewhat dangerous and Roman Reigns getting booed by 100.000 people while slugging it out with heel-but-not-really-heel Triple H.

2/18/2017

On the road to Wrestlemania!

It's Wrestlemania season!
Let us warm up with taking a look at how wrestling's biggest annual event has developed over time. We'll do this over multiple posts. In this first post we'll use very simple statistics regarding the number and duration of matches on the different Wrestlemania cards (all data comes from wikipedia.org).

Let's get some basic stats out of the way first:

  • A total of 311 matches have taken place at Wrestlemania over the last 30+ years (not counting matches on pre-Mania Sunday Night Heat or the Kickoff shows)
  • The number of matches on an average Wrestlemania card is 10 (9.7 to be precise)
  • A total of 3277 minutes (that's roughly 2 days and 7 hours) of bell-to-bell action since Wrestlemania I
  • The average Wrestlemania match takes 10 minutes and 34 seconds which is actually a bit more than I would have expected ...
which is probably thanks to ...
  • Longest match: Bret Hart vs. Shawn Michaels at Wrestlemania 12 (1996) with 1 hour and 1 minute
  • Shortest match: The Rock vs. Erick Rowan at Wrestlemania 32 (2016) with a whole of 6 seconds ... I like the Rock as much as the next guy but that was just plain stupid and made Rowan look like a total dork!
Now let's visualize how Wrestlemania has developed over time with respect to a) the average length of a Wrestlemania match in minutes (blue-ish line) and b) how many matches there were on each Wrestlemania card (yellow line).


Let's look at the key trends:
  • There is a clear trend to longer matches (on average) at Wrestlemania - at Wrestlemania 30 the average match took 16 minutes and 30 seconds compared to a meager 6 minutes at Wrestlemania 1.
    • The obvious outlier here is Wrestlemania 12 with Hart and Michaels totally boosting the average match length with their Ironman classic.
  • When it comes to the number of matches we see a peak in the 1980s and more or less random variation ever since. So the increased overall time that Wrestlemania takes up is used for longer matches on average, not more matches. That's a good development, if you ask me!

That's it for this post. Very basic still, but I'm saving some cooler stuff for later. ;-)

Wrestling and Analytics - weird but awesome!

Hey everybody, this it the first post in my Wrestling Analytics blog!

What this blog is about is data crunching and ... wrestling! Certainly an odd couple. However, we shall see whether - as is so often the case with wrestling (or, sports entertainment) - the stuff that I will cover isn't more interesting and entertaining than it has any right to be! ;-)

I'll try to regularly add interesting posts using methods of data analytics (which really only is a more fancy term for 'some form of statistics') and applying it to questions revolving around our favourite sport (and entertainment, I guess).

I am looking forward to your comments and feedback - let me know what you are interested in and I shall try to come up with some interesting (and maybe enlightening) numbers.