Posted by CoachSpud on 10/12/2016 12:35:00 AM (view original):
I don't think it is complex. I never said it is complex, just that it obviously isn't "basically random," as you claimed. But if you don't want to believe in it, well, then I guess it was randomness that Seble tweaked with his supernatural powers. You go ahead and claim whatever you need to claim to try to save face, it doesn't matter to me.

BTW, here is just one example of a formula for signing times (for any preference) that someone gave a few days ago in another thread:

“Seble described the signing process as each recruit having a % chance of signing in each cycle which will increase each cycle. So in the 3rd to last cycle, this particular player may have a 50% chance to sign. Then 2nd to last cycle he may have a 70% chance to sign that cycle. Then on the last cycle, it will be 100%. This is of course if he has a qualifying offer (scholly offer and has an interest level of at least High by 1 team).”

That is not “basically random.”

Posted by CoachSpud on 10/13/2016 1:28:00 PM (view original):

Posted by CoachSpud on 10/12/2016 12:35:00 AM (view original):
I don't think it is complex. I never said it is complex, just that it obviously isn't "basically random," as you claimed. But if you don't want to believe in it, well, then I guess it was randomness that Seble tweaked with his supernatural powers. You go ahead and claim whatever you need to claim to try to save face, it doesn't matter to me.

BTW, here is just one example of a formula for signing times (for any preference) that someone gave a few days ago in another thread:

“Seble described the signing process as each recruit having a % chance of signing in each cycle which will increase each cycle. So in the 3rd to last cycle, this particular player may have a 50% chance to sign. Then 2nd to last cycle he may have a 70% chance to sign that cycle. Then on the last cycle, it will be 100%. This is of course if he has a qualifying offer (scholly offer and has an interest level of at least High by 1 team).”

That is not “basically random.”

Yeah I posted that in a thread 2 hours ago. I was describing how signings for End of Period 1 preference works.

https://www.whatifsports.com/forums/Posts.aspx?topicID=495608&threadID=11042965#l_11042965

And I also posted in this thread yesterday that I never said all signings are random. I said Whenever is random and that it's a 50/50 shot of signing in 1st period or 2nd period. Which I still stand by.

... and which you've still never substantiated. You are entitled to your opinion. I hope you don't mislead newbies.

http://test.whatifsports.net/whatifsports/forums/Posts.aspx?topicID=420496&page=2

Whenever: Will now have about 50% chance of signing at some point in period 1 and 50% chance in period 2.

As Soon as Possible: Previously, these would sign almost definitely within 18 hours of the start of signings. I'm going to spread that out longer, so that they will definitely sign by the end of period 1, but not quite as quickly as they were.

By the end of period 1: Almost all will sign by the end of period 1, with odds increasing each cycle.

Late: None will sign in period 1, but almost all will sign by the end of period 2, with odds increasing each cycle.

Here's the change log for what it says for the signings. You can interpret this however way you want to Spud. Go for it. And all the newbies can make up their own minds about what this means too.

Fair enough. You can believe that is all there is to the formula, and there is nothing there to directly dispute that. I see where you're coming from.

setting aside the absolutely correct observation from Spud that one cannot know with complete confidence what "whenever" means without data on all the whenevers in several worlds......

given the limited information currently available, is the prevailing view that about half of whenevers will sign in period 1?

I had 18 whenevers in my recruit pool.
Only 2 signed in the 1st period.

I sent in a ticket about Whenever preference. Here's what I asked-

"In Beta, Seble said that there is a 50/50 chance of the recruit signing in either the 1st or 2nd session. However, after looking at the results from Allen and Phelan, it looks like only a very small % of recruit signings were listed as Whenever. Also, I noticed in the Dev Chat that there was a bug identified for the Whenever players' signing tendencies.

Was there a bug in Allen and Phelan or is this what we should expect from the Whenever signings?"

Here's the response -

"We've made some further adjustments to better get this to where we want it."

So there ya go. Pretty ambiguous. We'll have to wait and see what this means exactly by collecting data. I just looked at the Iba signings and 15 of the 78 signed top 100 players are Whenever. So something is definitely different from Allen and Phelan.

But also - 78 of the top 100 players have already signed and 1st session isn't even over yet. Wowza. But I know they're changing preferences to add more Late preferences to the better players so no need to read in to this too much. Just an observation.

I am pretty darn confused. Keeping one last team to see if I can enjoy the new game. The transition issues are a challenge - what words mean - and the EE problem. beyond these transition issues, I doubt that I want to work so much at a game - but that's just a subjective choice.

Dont need exact odds - some murkiness is fine - but right now I dont know enough about "whenever" to develop logical attempts at tactics

"We've made some further adjustments to better get this to where we want it."
So there ya go.

There ya go indeed. They either adjusted the laws of randomness, or they adjusted a formula.

or it is a formula that embeds random elements that are subject to adjustment?

They had said there was a BUG in the formula to determine if a Whenever guy would 'sign this turn' .. where guys were mostly going into the 2nd session.

This most likely means that they had the initial cutoff for 'signing this turn' too low, so the RNG happened and it was almsot always a NO until the last couple turns of the 2nd session.

Of course they use a formula (or equation, or programming statement if prefer) to:

1. Figure out the probability to 'sign this turn' and have a % sign and % not
2. Generate a random number (dice roll) for the event .. they get a number.
3. Do a lookup against the probability table to see if it is a sign or not.

If the lookup table was not properly built, it would render the RNG to not really random (or not a high enough chance to sign).

This is not an easy thing to program. It is not nuclear physics or rocket science .. but it is also not 2+2=4 easy either.

"This is not an easy thing to program."

So true, particularly since user behavior is so hard to predict and so central to this problem. If a lot of users put a lot of effort into Whenevers from the start, a high proportion of Whenevers become subject to possible signing earlier. If users hold off on putting effort into Whenevers, then many of them get pushed into later cycles at best, but not because they are programmed that way. This will all skew the time of signing, and it is likely to vary from season to season. Once WIS gets it pretty close to "about how they want it to work" they would be well advised to leave it alone and just say that "Whenever means whenever."

I am holding a grievance here. It screwed up my recruiting plans this first season, and now it is going to be changed. Great. Now a DII team is coming in on my redshirt hopeful because he was "whenever." No DI school should lose to a DII school if his preference is "strong conference." I have one other major grievance that we all know about, but I will save to after recruiting.

They just posted in critical news that whenever should be about 50/50.