Posted by cubcub113 on 5/10/2020 8:28:00 PM (view original):
Posted by Sportsbulls on 5/10/2020 8:23:00 PM (view original):
I feel preferences have a very large effect on how hard it is to unlock someone lol, I put the same amount of AP in many different players and the two correlations are overall ranking and preferences. I’m very confident those are the two leading factors. Team prestige as well obviously but we’re talking about different how long different players take with the same team so the prestige would be the same.
There is totally a 3rd factor which is randomness. Some players are just easier to unlock, others harder.
im starting to think there's no 3rd factor. or 2nd. that its only whatever builds position rating... which is, who knows what, it could include a random factor.
now that i think about it, i noticed when i started a year ago, a low a+ d1 school had some guys who were more that ~50 ap. for a higher a+, every single player is 50 ap ballpark max (and all like, 5* types, are in that range). but for the lower a+ it wasn't guaranteed - but outside the top 10, at each position, it seemed guaranteed. and it sort of seemed 'random' in there but i also wasn't trying to unlock the mystery or anything. it probably would have been position rank...
so let me try to state the hypothesis succinctly - each player is turned into a score in some mostly deterministic manner, and those scores are used to rank players for their position ranking, which is exposed to us. this position ranking isn't directly what is compared to the team's prestige, but rather, the score behind it. we may not know what causes the score (1 linear equation per position possibly plus a random factor??), but its possible that the behaviors are consistent relative to the score, for each position - at least for a whole season. the hypothesis is that there is a good chance that if the #150 PF requires ~100AP for some d2 school and the #151PF requires ~50, then every PF above 150 is at least 100 (modified by whatever, 5%, 12%, 18%, by preferences) to unlock while every PF under 152 is at most 50 AP to unlock (again, modified by preferences).
note - it is trivial for someone to try like 10 PFs with their d2 school and see if this hypothesis can be immediately disproven. its harder to prove, but if its false, it should be REALLY easy to demonstrate it. i don't have a d2 team, may need a volunteer...
hypothesis #2 - its possible the ranking scheme is a simple linear formula by position and if we had that information, we could generate the scores for players as needed, and through experimentation we could find the cutoffs on a score basis (which isn't impacted by seasonal recruit generation, like the position rankings are) for various levels of school (a+ d1, b+ d1, c+ d1.... a+ d2, a+ d3, etc) - and people could use those documented cutoffs to know, even for players who are unranked, a variety of things - the number of AP before small preference modifier - whether the player is likely to accept an inform of redshirt - and possibly, even stuff like whether the player will accept a redshirt once the season starts (or at least, what the probability is)?