Oh brother. I have no idea how that conventional wisdom exists but I wish I didn't have to spend so many posts every season trying to deflect it.
Granted, I realize at this point it will never succeed. There are people willing to swallow it throughout their life.
A pointspread is absolutely a prediction of the outcome. I got chief Nevada oddsmaker Michael Roxborough to emphasize that fact on The Stardust Line radio show in Las Vegas during the late '80s, while responding to my related question. Roxborough has class and will be straight forward, unlike many other Las Vegas sports betting big shots who think it is necessary and wonderful to fool the public and convince them something mysterious and ultra sophisticated is going on behind the scenes.
As always, the best way to verify is to point to the money line...the straight up odds, like -145 on the favorite and +125 on the underdog, which is what a 2.5 point game would hold. The pointspread comes first and then the money line is put up to correspond with the spread. There is a money line table that is immediately applied, to prevent subjective excess. So people wagering on the straight up outcome (money line) are doing so based on a number sourced from the pointspread, but people want to believe that spread itself has nothing to do with the projected outcome or "weighting the teams."
Pure genius
A sharp Las Vegas friend told me I should start using politics as an example. He might be correct. There is wagering on that also, and it is very similar to sports except instead of a pointspread there are political polls. The straight up odds on who will win is exactly the same as a sports money line.
For example, Hillary in the late going was roughly -300 and Trump +275, as mentioned in this link. BTW, the odds actually dropped later but this was a convenient link that I found immediately.
https://wgntv.com/2016/11/01/what-a...inton-winning-the-2016-presidential-election/
Those odds are sourced from the polling the same way a money line stems from the pointspread. Polling comes first and then the straight up money line is applied, albeit a political table and not sports based table. A 3 point theoretical edge in politics is greater than 3 point favorite in sports.
I don't think anyone would question that the poll meant that Hillary was favored to win. It is no different in sports except that power ratings supply the number instead of polling. Oddsmakers unabashedly steal power ratings from multiple sources and use a blended model for the consensus, not unlike 538 the political site takes polls and uses a blended number, like here:
https://projects.fivethirtyeight.com/trump-approval-ratings/?ex_cid=rrpromo
That blended consensus of stolen power ratings provides the pointspread. As someone else pointed out early in the thread, the Jets favored by 1 basically means that Miami entering tonight was considered roughly 2 points superior to the Jets. NFL home field advantage is 2.7 points but typically rounded up for convenience purposes.
Sigh. I realize this is more wordy than I intended. Probably not helping my cause. The basic problem always seems to be that people don't want to believe something as simple and mathematical as a power rating is responsible for all the pointspreads they see every week and all year long. They prefer the glamorized notion of Las Vegas wise guys subjectively analyzing each and every game, and brilliantly figuring out exactly what number will balance the action.
That is crap. The industry has no time for that. The pointspread as sourced from the power ratings does a fine job splitting the action. Then the books adjust the number based on action.
As always, this link depicts what happens better than anything else during an oddsmaker meeting. Notice how little discussion there is...maybe 30 seconds to one minute per game on average all year long. That's because everyone in the room is looking at the same thing...the blended power ratings:
http://vegasseven.com/2013/03/20/making-line/