Posted by CoachSpud on 10/10/2016 8:37:00 PM (view original):
" So about 2.5% of all players signed in the 1st session had a preference of 'whenever'. "

... which tells you nothing about how "whenever" is working. (a) It is too small a sample from which to draw any conclusion. (b) When you have data from full seasons in a number of worlds, you can see when the "whenever" guys actually signed. Then you can compare this to how WIS said it was intended to work. Then intelligent analysis of sufficient data will be possible.

I just posted full data from multiple worlds.

Posted by CoachSpud on 10/10/2016 8:52:00 PM (view original):

Posted by Benis on 10/10/2016 8:44:00 PM (view original):

Posted by CoachSpud on 10/10/2016 8:37:00 PM (view original):
" So about 2.5% of all players signed in the 1st session had a preference of 'whenever'. "

... which tells you nothing about how "whenever" is working. (a) It is too small a sample from which to draw any conclusion. (b) When you have data from full seasons in a number of worlds, you can see when the "whenever" guys actually signed. Then you can compare this to how WIS said it was intended to work. Then intelligent analysis of sufficient data will be possible.

I just posted full data from multiple worlds.

Full data from multiple worlds? Then I guess my computer connection was glitchy. It shows your post this way:
" I'll post the data I have from Phelan and Allen. Here are the players signed.

Allen

	Signed	Whenever
D1	117	3
D2	28	2
Total	145	5

Phelan

	Signed	Whenever
D1	73	1
D2	24	0
	97	1

Combined

	Signed	Whenever
D1	190	4
D2	52	2
	242	6

Your answer appeared to involve only D1 and D2, and 12 signed recruits with preference of "whenever" in three worlds. I'd be interested in full data from full seasons from multiple worlds so intelligent analysis is possible, and I'll wait to see that data and the comparison to the way WIS intends it to work.

That's the data from the 1st session from 2 worlds. If someone wants to post all the D3 signings thatd be great too. How about you post some data for your world and try to actually contribute something.

And almost 250 signings is a small sample size? Includes the entire top 100 recruit signings for 2 worlds.

Okay how about you ask the WIS execs when you guys go golfing this weekend. Report back to us on Monday.

Wow Spud, you are a piece of work. Do you actually know anything about statistics?

With 100 whenevers and less than 5% signed in cycle one, you can say with 99.99% certaintly that Whenever is not an equal chance between signing in cycle one or cycle two.

Posted by CoachSpud on 10/10/2016 10:58:00 PM (view original):
With 1000-some odd recruits per world per season, and 10 worlds, I am not the least impressed by your undocumented 100 whenevers. I said what I would look at to find enough data for intelligent analysis, and 12, or 100, isn't even close. Wake me when there is enough data for intelligent analysis.

No one forced you to comment on this thread. You're welcome to stay asleep.

Posted by CoachSpud on 10/10/2016 10:58:00 PM (view original):
With 1000-some odd recruits per world per season, and 10 worlds, I am not the least impressed by your undocumented 100 whenevers. I said what I would look at to find enough data for intelligent analysis, and 12, or 100, isn't even close. Wake me when there is enough data for intelligent analysis.

Just ran the stats analysis myself. With the current sample sizes, as reinsel stated, there's over a 99.5% chance that "whenever" is biased towards Late. So either "whenever" doesn't mean "whenever" and should be renamed, or it's not working as intended from the Dev chat. Sampling - it's a real thing.

Posted by CoachSpud on 10/10/2016 11:12:00 PM (view original):
Compare that to how WIS says it works, and then see if you can see your problem.

Oh. We know what our problem is. Not too difficult to pinpoint.

You tell 'em Spud hole!!

Whoooosh!

Right over your heads.

Lovin' it. And look at all the exercise you little guys are getting trolling me.

Posted by reinsel on 10/10/2016 10:25:00 PM (view original):
Wow Spud, you are a piece of work. Do you actually know anything about statistics?

With 100 whenevers and less than 5% signed in cycle one, you can say with 99.99% certaintly that Whenever is not an equal chance between signing in cycle one or cycle two.

Agreed. They can sign in the early period, but most (90-95%) wait.

If you want Early or before the End of Session 1 recruits, you can pick that.

Whenever means basically no preference. In practice it means .. most likely not early (it would seem).

The amount of effort does not impact signing (of course, NO offer means no signing). But if someone wants to sign early or before the end of the first session, someone still needs to be Very High to sign them.

Yeah Hughesjr, I am in agreement with your analysis.

In practice whenever seems to mean .. maybe 95% late and 5% early.