I said he’s not wrong and the question I asked was if the NET starts out with a predetermined rating.
Regarding the pre-determined rating, some background before addressing NET in #2 below.
KenPom is perhaps the most respected predictive team ranking forumula out there -- hell Vegas moved its gambling odds years back to metrics that more closely followed KP.
KenPom ratings don't start at "scratch" at the beginning of the year. He uses some formula looking at prior year play and what is coming on. Its not 100% accurate, but once 5 to 7 games are played. the current year data is connected enough that the impact of pre-season rankings is becoming quite irrelevant. And by the time OOC starts, preseason rankings zero impact - all data is connected on current year play. It would be the same for BPI which is part of NET.
#1) The RPI had no predetermined rating - it started from scratch each year, which created some insanity early on until teams and data started connecting more.
#2) The NET would probably have some pre-determined element to start the year, because it uses BPI as part of its formula (a predictive power metric like KP). BPI, like KP. does have a formula to rate teams to start the year that would wash it outself out as current year data connects. But the predictive power system metrics aren't dominating the NET formula - we still see some early nonsense in the NET. like the RPI, when data is not connected enough.
At the end of the day there really is no pre-determined factor here that influences anything. The BIG, the BIG 12, the SEC all had the best NET's as a conference, and they also clearly had the best OOC results.
-------
But in terms of what he said.
a) What exactly was he right about? Play well as a group or play like **** the NET will respond to you in either direction.
It can be argued it does it too much for those within the P6 conferences that do really well OOC and vice versa hurts those too much that do poorly. But still its not the NET that's killing you. Its conference performance.
b) He claimed we should use more eye test. In theory the "eye test" argument always sounds great, but its so subjective, impossible to implement fairly, and let's be honest about when the eye test is used. The eye test argument is only used when objective data ... actual W's and L's, don't support your case.
c) He said something about the 20 game schedule which was absoluite nonsense. If your conference is strong or on par with the others, the 20 game schedule is great. It gives you more quality games and win opportunities.
d) I don't think final conclusions should be done on how your top teams do in the tournament. Can't be ignored (its part of the equation), but it certain;y can't be overemphasized either. Lots of fairly random stuff happens in a one and done.