(previous years performance/strength) that isn’t part of what it’s meant to measure (this year’s performance/strength).

It would be better to use the very limited data you have and then it would be easy to give the conclusions only as much authority as the amount of data warrants.

Early season ranking would shuffle dramatically in the first few weeks, which would be much better because in addition to removing unearned bias, it would demonstrate how meaningless they are.

As is, they are given way too much credit.

If WSU is ranked higher than UW early in the season, great. It doesn’t matter if they will or won’t end up that way at the end, because that will be demonstrated by then. But rankings that are supposed to order teams this year should be based on what teams do (or have done) this year.

If you think you don’t have enough data, the answer is wait to collect more data, not insert data from outside the set or arbitrarily base it on what you think will happen.