I have to go out. I will posts the results including totals when I get back. i will do it total for total, and then you can see that is has absolutely no correlation.
Daringly, it all depends on what lines you are using. I wil in fact cite the database that I have that is WORSt case scenarion in regards to MY argument. In other words, I will use the numbers that would best go to prove YOU are right. And STILL they won't be close.
I could also use a couple other databases that have more favorable numbers to MY argument, and then it REALLY shows that it is meaningless.
These arguments are all theory anyways, because for the most part to "PROVE" it you HAVE to use flawed data. Flawed as in using games that closed -3 and landed on -3. For almost HALF those examples of a push, which is your "evidence" to say they hit 10.5 or 8.4 or 9.6 percent of the time, another number was available elsewhere or at a different time. So when you say there is X amount of this happening and that happening, you are using flawed results, just like I am. But since it is the only way we have to determine probability we have to use it.
Like I said I have to go out, but I will throw up my numbers for every total(which again will be a judgement call since they will be varied as well) but we can get a range to work with.
I have been arguing this debate for 10 years, trust me I have the result down, and almost memorized by now.
The 3 is the most vastly overrated angle/number there is in sportsbetting. Simply because people have been programmed to believe it.