MMAPayout.com is pleased to announce it will be introducing the newest version of its PPV predictor in next week’s UFC 104: Payout Perspective.
This version of the PPV predictor will make use of regression analysis across two different variables (Spike’s Countdown to UFC viewership & UFC live-gate revenue) in order to come up with an initial, statistically-supported estimate of UFC buyrates. Then, we’ll supplement the regression figure with additional qualitative factors that might have influenced the buyrate one way or another.
Regression analysis is basically just a method used to identify a pattern between a set of related data sets: e.g., the relationship between the number of viewers that a Countdown show receives and the number of buys the PPV receives. It allows us to find a line of “best fit” through the data that we can then use to create a formula, which will allow us to enter the Countdown viewership and predict the PPV buyrate.
The variables themselves have proven to have a solid, positive correlation with PPV buyrates, given that the current sample size is quite limited (anywhere from 30 to 45 depending on the input variable).
Please Note:
The figures used in the regression analysis are, themselves, estimates. Expecting the formula to be 100% correct is unwise.
What’s the point, then? There are several:
1.) It’ll still be the best – and only statistically-supported – analysis tool on the web.
2.) The regression analysis is just another means of slowly piecing together what does and does not impact PPV buyrates – it’s essentially an on-going study of the business. My hope is that we continue to learn more and more as we go along.
3.) It’s a point of discussion, and MMAPayout.com openly encourages its readers to share their opinion given the various different factors that might influence the buyrate (quantitative or qualitative).
I would strongly suggest also throwing google trends into the mix. Google trends is pretty amazing at showing popularity and good be another key element to predicting PPV numbers.
http://google.com/trends?q=UFC&ctab=0&geo=all&date=all&sort=0
Google is almost worthless IMO. PPV numbers often depend on how many hardcore choose to buy the event. Hardcore people are not googling UFC 104. There is no need. Harcore fans know the info or knew where to find it. Some noobs may google. But there is no way to know how many will buy the PPV. How many people googled it mulitple times?
Quite simply you dont know if the high google searches are due to increased interest in the PPV or the PPV is full on unknowns so people have to google to find the info. Id be more inclined to google an event i didnt know anything about rather than a UFC 100 that everyone already knows about.
According to google trends. Kimbo Slice should draw the same PPV buys as Brock Lesnar.
http://google.com/trends?q=Kimbo+Slice%2C+Brock+Lesnar&ctab=0&geo=all&date=all&sort=0
That’s the beauty of regression. Plug in the variable and see if it improves the predictive power of the equation. If not, leave it out. The MMA Buzz ratings or change in buzz rating leading up to an event may be an interesting variable to test as well.
Payout, do you mind sharing your r2 for viewership and live gate variables?
If Kimbo was in the UFC and fighting guys like Randy Couture I’m quite sure he would have gotten the same ppv numbers or better than Brock. Let’s be honest the guy is a smash hit when it comes to rating’s so that certainly isn’t shocking.
Brian I get what your saying but one thing that is very clear when you look at google trends is the fact that it shows the buzz around an invent. For instance if you type in 100, 101, 102, 103 you can directly tell which ones were the most popular. You can literally just look at the graph and tell which events had the higher and lower ppv numbers.
You can’t use google trends to get exact numbers but what you can do is look at the hype surrounding the event and compare it to others of that same year. I’m not saying to use it as the sole tool I’m just saying it can definitely be beneficial when supplementing what is already in use.
Google can be useful but not more or different than the tools already used. If offers nothing new to help the estimates. I have always been good at guessing PPV numbers just by the fighters past draws, the recent UFC Buys and the amount of press the event gets plus guess on other factors. But just like using The count down shows to estimate PPVs it only gets you in the ball park and even then there are events that greatly contradict the Countdown ratings.
Google is just another tool to get you close 3 -5 events then leave you scratching your head just like i believe) UFC 100 count down ratings did.
I agree with you Brian, just like I said it is good when suplementing the other factors. IMO it’s the best indicator of showing the hype factor.