I think you are getting confused w some public PRs sites may provide online like you say once a week vs PRs cappers develop on their own to use when analyzing games.
And you adjust these personal PRs daily based on injuries, weather and all other news. Joe Burrow is out, you adjust your PR down by say 6 pts etc, a CB is back from IR + 1 pt etc etc.
Hope that helps and again it is just another tool, some use them some don't.
While your theory sounds like what any person would expect, - its radically wrong again even assuming a 6pt drop in losing QB Burrow.
Just the fact that Cincy actually won the game SU vs Jags as a huge dog, exposes that any power rating used is speculative at best.
So in the first week The Bengals lost without Burrow to Baltimore when he went down early in the game, you wanted to adjust the rating by 6pts for the next Gaame.
The first full game without him lost to Pittsburgh. So Its for sure that you could not move the rating number back up after the loss, but would be even more susceptible to move it down again.
Therefore, again - using that Pittsburgh loss as a way to reduce the PR, as noted in week-by-week results.
Technically now, your "adjustment would have been another 2-3 pts downward when adding in the loss of that first game without burrow.. So NOW your PR has dropped at least 8-9 pts and all od a sudden youre betting against the Bengals in the next game with a substantial drop in the rating which is indicating to you that Cincy is far worse without Burrow.
Then all of a sudden BOOM. The Bengals as a 10pt dog, stun Jacksonville who's PR was way higher than the Bengals at game time - and you are so far off on the PR number - that you cant recover using them again to find an advantage in the spread. Just the fact that anyone would use a 6pt move in the expectation of a line advantage, is when these numbers are skewed so badly.
News reports, injuries an other factors that you may feel are worth using to adjust a Rating, is when when the first number created is wrong, it then compounds the power ratings even more wrongly and the lag time I mentioned, is very real now and it causes many to misread the changes radically, into a false advantages for or against any team involved in the next game.
Notice now that the line has been radically adjusted for Cincy now on the next game but your PR numbers are at least 2 weeks behind and are way off being anywhere near accurate, in any way.