Posted by buddhagamer on 8/1/2019 1:21:00 AM (view original):
Preferences are multipliers (not added bonuses). Each cycle, all recruiting effort is converted into recruiting credit (a calculated number) which is added to your effort total. Subtracting from that total when promises are withdrawn can be easily computed (since you know what the multiplier was prior and how much effort in each cycle). Seble then applies a large penalty for withdrawing of minutes (unless you offer more in that same cycle) or for start to the point when reaching for a higher recruit it makes it near impossible to overcome.
One flaw in shoe's example is each cycle, your preference may NOT be identical in some cases. Strong Defense for example will change game to game so 60 AP in once cycle vs. 10 over 6 might yield different totals. Many coaches make the mistake thinking all preference grades are identical (i.e all VG are not created equal... everyone has a preference number where Seble is just showing a "grade" when you score in particular ranges).
the first part of what you are saying, i think you are just over simplifying it a bit (or maybe a lot) in your head, about the implementation cost. sure, if you have the effort per cycle, and the multiplier per cycle that was in place at that time, you can compute cycle by cycle and back that out when minutes promises change. however, that involves tracking historical state in ways that are otherwise not needed, and also forces you to think about the order of operations in ways that are not otherwise needed (do you apply the promises bonus first or last, etc). there are edge cases and complications all over the place with the current recruiting scheme, particularly around the ability to withdraw things. there have always been bugs and issues with withdrawing actions, whether those are automatic or manual withdrawals, in HD recruiting, and there are in today's game, too.
long story short, the above is not simple, it is not clean, and building that kind of logic is inherently bug-prone. my guess is part of the reason seble applies such a penalty for backing out promises, is to ensure there is no way there is a net-positive loop possible where you continuously apply and withdraw promises - and without a superficial additional penalty, its very likely there would be a bug in seble's logic and tracking somewhere, that would allow such a positive loop to exist. i can already guarantee you there is at least 1 positive loop present in recruiting today, not all that different from how in 2.0 i discovered you could exploit withdrawl logic to get free evaluation visits - just only the informational part of it, not the credit part of it - which all in all, was the only reason people did evals at all. of course i reported that, and will do the same in 3.0 when i get a bit further. or how in 2.0 everyone knew you could pull promises by redshirting but could immediately wipe that out by withdrawing the redshirt. it wasn't a positive effort cycle or anything, but it got around the intended design because there was no check on the automated withdrawl of promises caused by redshirting, no check in place to insure you wouldn't just back out the redshirt and sort of glitch your way out of promises.
i really had zero intention of looking for those glitches that early, i've only been back 2 months, and i only look for stuff like that to report it and reduce the odds coaching i'm competing against are using those exploits against me. but i stumbled by pure accident into a free effort scenario already, so its sort of moved up my time table.
this is just one of those things, where it seems simple if you haven't had to develop anything like it before, and aren't intimately aware of how bug-prone this kind of stuff can be. with a software dev background, i know better - which is why i know to look for those withdrawl and order of operations type exploits in recruiting, and have ultimately found multiple which IMO sort of proves my point. also, seble was well aware of these issues and worked pretty damn hard to avoid them in 3.0, which kind of just goes to show, its not nearly as simple as you are making it out to be.