First the spread matters in every game...that is why they have them...No clue where you got your numbers but here are the #'s as follows
Dogs are 57-32-2 ATS this year
Dogs winning S/U are 40-51...that is a 17 game difference this year alone.
Perhaps there is a misunderstanding, but wizardofodd is correct in his presentation that the point spread matteres only 17% of the time IN THE NFL . . . Not college. So far this year it is around 18% but over many many years it has stabilized closer to 17%.
Nobody is talking about your data which you state as dogs are 57-32-2ATS, etc. He is not talking about spread records and that may be the source of confusion. Let me try to explain with a few different (similar meaning) words.
When you retrospectively analyze the outcomes of all NFL games INCLUDING PLAYOFFS, the spread does not matter in the outcome approximately 83% of the time. This means that if you can pick the WINNER of the game, make the bet and you will be cashing your tickets at an 83% clip. Conversely, you will lose those same bets 17% of the time. This is not a debatable topic. It is proven fact for many, many years. The only point of debate is whether someone understands the concept. This is not ATS.
Once again, here it is stated again in yet anothe way using this weekend's games as an example: GB -5.5 at St. Louis - if you think GB is going to win, lay the 5.5 as the number is not likely to come into play (Once again, 83% of the time if GB wins, they will win by more than 5.5). Using a different game as another example, let's look at Minnesota -6.5 at home playing AZ. If you have an opinion on this game and you believe AZ will beat MINN then play the ML on Arizona because we know that the point spread only will come into play at approximately 17% of the time. If you fell that MINN will win the game, then lay the 6.5 and don' go buying down to 6, etc.
If you don't agree with this, here is how you can prove it to yourself although it has been done for you precisely for weeks 1-6 by wizardofodd. And it has been done for decades by others. When week 7 is complete, go through the NFL results and ask yourself this simple question. Did the winner of the game cover the spread? Simple as that. Did the winner of the game cover the spread? If they were a dog and won outright, they covered the spread. If they were a favorite and won by more than the closing Vegas line, they covered the spread. Simple as that.
The only exception to this format is that DOUBLE DIGIT POINT SPREADS DO NOT APPLY TO THIS CONCEPT.
So when you check the results on Monday after the games have been played, you should find that approximately 17% of the time the point spread mattered in the outcome.