This is not my system. It was created by a pro handicapper using some type of algorithm. As far as I can tell, it's been around for at least a couple years. The success rate has varied.
When a opening total crosses a number that can be divided by 7 or 10 to produce an whole number (meaning no decimals), then it becomes a key indicator on which side of the total you should play. These indicators work under the same premise of when an ATS line crosses 3 and 7, indicating which way you should lean.
For example, total opens at 48.5 and crosses 49 to 49.5 .
49 / 7 =7
49.5 / 7 or 10 = 7.07 or 4.95
in this example, playing the under would be proper. Making your selection opposite of the line movement when it crosses one of the key numbers.
Key numbers that can be divided by 7 or 10 to make a whole number = 30, 35, 40, 42, 49, 50, 56, 60, 63, 70,77
The reason numbers divisible by 7 or 10 are key is because these are the average number of point leads/consecutive scores on alternating possessions that normal teams make throughout game progressions.
If the movement crosses two Key numbers (ex. Open at 43 and drops to 39.5) then it's a very strong indicator of making the play
If the total starts at a key number, wait for it to pass another key number before making a play.
As always, take into account match-up, weather, injuries, win/loss records, etc.
It's suggested to track the movement early and make decisions as soon as it crosses the key number to take advantage of preventing further movement the other way. The point of catching it as soon as it crosses the key number is to lock it in at a low or high number depending which side you take based on the movement. If you don't monitor it, it puts you into a spot where you shouldn't make a play based on this method.
I'm not sure if this works on spreads. I've only seen it used for point totals.
This is for the NFL only. Why doesn't it work for college totals?
College football has a significant variance vs NFL in the frequency of scoring patterns/great or poor play/unpredictability due to the nature of who's playing (college kids,conference quality, etc).
A significant percentage of college games are also very high scoring, which skews favor onto over bets. The purpose of this model is to eliminate the human element of favoring one side and making the play based on math, fundamentals, and taking advantage of the line movement on the total.
NFL ending totals and scoring sequences have remained quite steady throughout the years. And most NFL totals will always be close to or right at a number which is one of the key numbers. Crossing the key number quickly is the ultimate indicator, and in a college total you may never even approach one.
It is fairly uncommon to find college games with a total less then 50, as most of them creep up into mid 50's or higher. You have more chances in a 40's type total to cross multiple key numbers.