Posted by rmdlwa on 1/1/2020 6:13:00 AM (view original):
OK so I just had my first player bolt - he was a Junior and I was super excited to have him back for a nice run next year. He emailed telling me he was leaving for the draft but it appears the draft just took place and I don't think he was drafted - what happens next? He appears on my roster everywhere except the screen where you recruit - he is gone from that roster and I just signed a recruit in his place but am waiting for the roster updates...any chance I get him back?
There is a little bug when you view draft results. It defaults showing you last seasons draft so you need to change get the season manually.
1/1/2020 2:01 PM
Posted by tgblackw on 3/31/2019 2:48:00 PM (view original):
Hey fellas,

I'm relatively new to D1 and have never had an EE. However, I have a sophomore who is # 18 on the big board and is "on the fence". What are my chances of getting him back? What kind of prestige bump would I be looking at if he were to leave?

Here he is:

https://www.whatifsports.com/hd/PlayerProfile/Ratings.aspx?tid=0&pid=4135377
That's great man and you probably got lucky.
1/1/2020 2:04 PM
some good info in here, but would like to clarify and add a few things -

1) summary of EE board - think of the board as simply being a range, a gradient of odds to leave. there are 3 ranges to concern ourselves with - assuming 'staying' means 0% to leave (which i believe to be true):
'likely staying' - about up to 30% or so, could be more in the 20s - 30s actually, its more that i think a 100 ranked junior is more around 30% i guess
'on the fence' - from about 30% to about 70%, maybe 65% (if on the fence is 20-65, i think juniors at 100 start more like 30%)
'likely going' - from about 70% to about 95%, the really high guys come back decently often but it does seem like its more 4-8 than 1&2

for whatever reason, the bottom half or so (i usually think 100-40 or so but that is just off observation, benis's 50 number is very well better being based on stats) have similar (low) odds, but then it really picks up on the way to 28, where player's move up a range.

there are 2 key cutoff numbers - 28 and 18, and these apply to everyone (there is also a cutoff for freshman, they start on the fence at 17 and go to likely going somewhere, maybe 8, but i kinda forget and its super rare).

so in the end if kinda looks like this
[100 ------------------------------------- 50 / 40 ------------------------------------ 28 ][ 27 -------------------------- 18 ][ 17 --------------------------- 0 ]

alright the formatting worked incredibly awfully so i give up - just typing it out
100-28: jrs are roughly 30-70% in here, it really picks up around 40/50. sophs are roughly 10 or 20% on the low end, up to 20 to 30% on the high end. fr are 100% to stay
27-18: jrs are roughly 70% on up in here, probably up to 80% or so. sophs are roughly 30-70%. fr are roughly 20-30 to 70%
17-1: jrs are up to a good 95% on the high end (4-8 seem to stay decently often but not 1-3 so much), sophs are probably 70-90%. fr 30-90%. there is a cutoff for fr in there from on the fence to likely going but i forget where, maybe 8? its super rare.

benis could probably slightly improve on these % numbers, he had some really good stats he ran, but that is the general model it follows at least.

2) every draft pick is worth a prestige bump, but its a fraction of a partial grade for every player. the #1 overall pick is worth substantially less than 1/3rd of a grade (one partial grade). whether you go up a grade or not usually is a function of how close to that next grade you were, to start with. nobody has ever done a study but i can lay out an approximate range.

the top of the 1st draft guys are worth significantly more than others. i would estimate the top 10 picks are on average worth about a quarter to a third of a partial grade (so about a tenth of a full grade). second round picks are worth decently less, probably like 1/10th of a partial grade on average, maybe a bit more (.15?). it is possible for your displayed grade to go up 2 partial grades (a- to a+) but it takes some ridiculous **** like 5 first round picks.

3) this bump should last for exactly 1 season, well it lasted for 1 season in 2.0 but i have no idea what 'one season' is anymore, in 3.0 with its two sessions and crazy schedule and the whole question about whether prestige changes even impact session 2. overall, this bump does not come *close* to salving the wound that is an EE. it didn't in 2.0 and EEs were half as bad then.

4) the year of the player impacts how likely the are to leave, for sure - the 25 ranked soph is way less likely to go than the 25 ranked junior, and so on for every position. however, the listed labels, in my opinion, are for absolute ranges and do not take into account class. i believe the 'cutoff' between say likely staying and on the fence, is a fixed % - whatever that is, let's say 30%, its 30% for juniors, sophs, and freshman.

so what benis is saying, on the fence sophs more likely to leave - its sort of right but misleading. of the set of 'on the fence' sophs, a higher % will leave than 'on the fence' juniors. however, the only reason for this is that SO many juniors exist in that 50-100 range where they are all on the lower end of the 'on the fence' spectrum. a junior can't be listed without being at least 'on the fence'. meanwhile, on the fence sophs only start at 27, and they cut off at 18. so, only a couple of the sophs are on the very low end of the 'on the fence' range, because they are so quickly getting up to the high end of that on the fence range, just over the span of 10 spots on the rankings. roughly, the rate at which sophs leave should bisect the on the fence range (if its 30-60, sophs should leave at 45% on average, more or less). the bulk of juniors leave near the bottom of the range, 30-40%, which pulls down the average.

but 'on the fence', whatever that means, is the same for all classes.

5) the board is basically the be-all and end-all for regular season rankings. its not some approximation, its the actual standings the game will use. the final rankings are only impacted by 2 things (beyond the board rankings themselves) - post season success (this is purely team based, champions should consider all their players to be a dozen or so spots higher than they are), and national all american awards (which are decently rare in big conferences where most EEs come from; its usually good players in garbage confs who are just juicing the stats and the game is dumb)
1/2/2020 4:40 AM (edited)
so, to the OP, fd was right, this guy is a soph on the highest spot for on the fence (18), so hes probably in that 60-70% to go.

how much prestige bump? it depends on when he actually gets picked, but at 18 most of the guys are leaving, so if hes 17, i would roughly guess 0.2 of a partial grade (0.67 of a full grade - so like going from a 95 to a 95.67 for 1 season). could be up to 0.25 of a partial grade, but i'd be shocked if it was much higher (no way its over 0.3)

your post season success (and AA awards of him/players around him) could impact it too, but sort of impossible to project that without knowing that info
1/2/2020 4:14 AM (edited)
summary of EE board - think of the board as simply being a range, a gradient of odds to leave. there are 3 ranges to concern ourselves with - assuming 'staying' means 0% to leave (which i believe to be true):
For the data I collected, 3 out of 125 players that were listed as Staying actually left early. So pretty much 0% but not exactly.

'likely staying' - about up to 30% or so, could be more in the 20s - 30s actually, its more that i think a 100 ranked junior is more around 30% i guess
'on the fence' - from about 30% to about 70%, maybe 65% (if on the fence is 20-65, i think juniors at 100 start more like 30%)
'likely going' - from about 70% to about 95%, the really high guys come back decently often but it does seem like its more 4-8 than 1&2

Likely staying is ~15% chance to leave early
On the Fence is ~40% for JR and ~60% for Soph.
Likely Going is ~87% for both Jr and Soph.

1/2/2020 2:56 PM
"so, to the OP, fd was right, this guy is a soph on the highest spot for on the fence (18), so hes probably in that 60-70% to go."

Out of the 49 players I tracked that were Soph ranked between 18-27 (On the Fence), 29 left early for a 59% rate.
1/2/2020 3:01 PM
Posted by Benis on 1/2/2020 2:56:00 PM (view original):
summary of EE board - think of the board as simply being a range, a gradient of odds to leave. there are 3 ranges to concern ourselves with - assuming 'staying' means 0% to leave (which i believe to be true):
For the data I collected, 3 out of 125 players that were listed as Staying actually left early. So pretty much 0% but not exactly.

'likely staying' - about up to 30% or so, could be more in the 20s - 30s actually, its more that i think a 100 ranked junior is more around 30% i guess
'on the fence' - from about 30% to about 70%, maybe 65% (if on the fence is 20-65, i think juniors at 100 start more like 30%)
'likely going' - from about 70% to about 95%, the really high guys come back decently often but it does seem like its more 4-8 than 1&2

Likely staying is ~15% chance to leave early
On the Fence is ~40% for JR and ~60% for Soph.
Likely Going is ~87% for both Jr and Soph.

I'll add that within the Likely Going field, it does matter if the player is ranked 5th or 15th.

For example, a JR ranked in top 5 has a 95% chance of leaving early while a Jr ranked at 15th has only an 80% chance of leaving early - even though both are listed as Likely Going.
1/2/2020 3:04 PM
Posted by Benis on 1/2/2020 2:56:00 PM (view original):
summary of EE board - think of the board as simply being a range, a gradient of odds to leave. there are 3 ranges to concern ourselves with - assuming 'staying' means 0% to leave (which i believe to be true):
For the data I collected, 3 out of 125 players that were listed as Staying actually left early. So pretty much 0% but not exactly.

'likely staying' - about up to 30% or so, could be more in the 20s - 30s actually, its more that i think a 100 ranked junior is more around 30% i guess
'on the fence' - from about 30% to about 70%, maybe 65% (if on the fence is 20-65, i think juniors at 100 start more like 30%)
'likely going' - from about 70% to about 95%, the really high guys come back decently often but it does seem like its more 4-8 than 1&2

Likely staying is ~15% chance to leave early
On the Fence is ~40% for JR and ~60% for Soph.
Likely Going is ~87% for both Jr and Soph.

thanks for the info! pretty awesome stuff.

did you pull the national awards and more importantly post season success of those 3 teams who lost guys 'staying'? my guess is you missed the post season part (intentionally, but still). teams with freshman on the board typically are pretty good teams :) there is absolutely no question that a team can do well in the post season, and a staying guy can move up enough to be a non-0 chance of leaving. but that doesn't mean staying, as a starting point, doesn't mean 0% - i think? (as in, if you stay in the same spot after the post season and awards adjustments, you are 0%).

the on the fence, i maintain my prior explanation. you are just dealing with lots of juniors in the 30-40% range in the 100-40 range of the board. with a relatively small number in the 40-28 range where they ramp up rapidly to 'likely going'. meanwhile only 18-27 sophs are 'likely going' so if the range is 30-70 they'd be 50% or so. your 60% is higher. but i don't know, i seriously question the sample size you have of sophmores who are on the fence, given how few there are.

that likely going is higher than i expected though, as an average. i'd like to mull that over, there's a couple things that come to mind that could be at play there. i look at the start of 'likely going' as when my 40th on the board junior wins a title and then leaves at some rate. its very hard for me to evaluate a just-at-28 guys because 1) i plan to well to have more than 1 of those every blue moon and 2) that wouldnt mean **** because that team likely would have a good run and he'd be more like an 18 guy. i feel like 87% is without question much higher than the true odds of leaving as the 27th junior. maybe the range goes to 95% quick, everyone in top 10 is at least 95%, allowing the 27th guy to be a bit lower. im not sure.

you have more data on that likely going one? that is the one that sort of does not really jive with my prior expectations
1/2/2020 7:19 PM (edited)
one interesting thing that i never consider was shoe's suggestion about the model having a guy 'move up a spot when the guy above him stays'. i never thought about it like that, i kinda don't think it works that way but it would be REALLY hard to detect the odds impact, so i don't feel like i can be confident.

i mean, if the #5 guy decides to come back, the #6 guy would obviously become the #5 pick. but do his decision odds change? feels like it could just as easily be written either way. that is pretty interesting. we could probably determine it with some experiment. i always just assumed the game dealt with guys not leaving by taking folks below 100, not by shifting the odds - and it feels like the cutoff does move significantly season over season. more than it should if the game was giving the higher odds to the next man up, so to speak? no idea.

hey benis, buddy, could you figure that one out for us?
1/2/2020 7:30 PM
Posted by gillispie1 on 1/2/2020 7:19:00 PM (view original):
Posted by Benis on 1/2/2020 2:56:00 PM (view original):
summary of EE board - think of the board as simply being a range, a gradient of odds to leave. there are 3 ranges to concern ourselves with - assuming 'staying' means 0% to leave (which i believe to be true):
For the data I collected, 3 out of 125 players that were listed as Staying actually left early. So pretty much 0% but not exactly.

'likely staying' - about up to 30% or so, could be more in the 20s - 30s actually, its more that i think a 100 ranked junior is more around 30% i guess
'on the fence' - from about 30% to about 70%, maybe 65% (if on the fence is 20-65, i think juniors at 100 start more like 30%)
'likely going' - from about 70% to about 95%, the really high guys come back decently often but it does seem like its more 4-8 than 1&2

Likely staying is ~15% chance to leave early
On the Fence is ~40% for JR and ~60% for Soph.
Likely Going is ~87% for both Jr and Soph.

thanks for the info! pretty awesome stuff.

did you pull the national awards and more importantly post season success of those 3 teams who lost guys 'staying'? my guess is you missed the post season part (intentionally, but still). teams with freshman on the board typically are pretty good teams :) there is absolutely no question that a team can do well in the post season, and a staying guy can move up enough to be a non-0 chance of leaving. but that doesn't mean staying, as a starting point, doesn't mean 0% - i think? (as in, if you stay in the same spot after the post season and awards adjustments, you are 0%).

the on the fence, i maintain my prior explanation. you are just dealing with lots of juniors in the 30-40% range in the 100-40 range of the board. with a relatively small number in the 40-28 range where they ramp up rapidly to 'likely going'. meanwhile only 18-27 sophs are 'likely going' so if the range is 30-70 they'd be 50% or so. your 60% is higher. but i don't know, i seriously question the sample size you have of sophmores who are on the fence, given how few there are.

that likely going is higher than i expected though, as an average. i'd like to mull that over, there's a couple things that come to mind that could be at play there. i look at the start of 'likely going' as when my 40th on the board junior wins a title and then leaves at some rate. its very hard for me to evaluate a just-at-28 guys because 1) i plan to well to have more than 1 of those every blue moon and 2) that wouldnt mean **** because that team likely would have a good run and he'd be more like an 18 guy. i feel like 87% is without question much higher than the true odds of leaving as the 27th junior. maybe the range goes to 95% quick, everyone in top 10 is at least 95%, allowing the 27th guy to be a bit lower. im not sure.

you have more data on that likely going one? that is the one that sort of does not really jive with my prior expectations
Correction - only two guys listed as Staying actually went early

First guy was ranked #31 on Big Board. His team went to Elite 8 and had no national awards.
2nd guy was ranked #37 on Big Board. His team went to Sweet 16 and had no national awards
1/3/2020 11:36 AM
Left Early Likely Going
Big Board 1-5 6-10 11-17 18-27
JR 95% 98% 83% 78%
Left Early Likely Going
Big Board 1-5 6-10 11-17
SO 89% 91% 83%
1/3/2020 11:38 AM
For the sample size
Left Early Likely Going
Big Board 1-5 6-10 11-17 18-27
JR 59/62 61/62 65/78 78/100
Left Early Likely Going
Big Board 1-5 6-10 11-17
SO 24/27 20/22 25/30
1/3/2020 11:39 AM
Posted by gillispie1 on 1/2/2020 7:30:00 PM (view original):
one interesting thing that i never consider was shoe's suggestion about the model having a guy 'move up a spot when the guy above him stays'. i never thought about it like that, i kinda don't think it works that way but it would be REALLY hard to detect the odds impact, so i don't feel like i can be confident.

i mean, if the #5 guy decides to come back, the #6 guy would obviously become the #5 pick. but do his decision odds change? feels like it could just as easily be written either way. that is pretty interesting. we could probably determine it with some experiment. i always just assumed the game dealt with guys not leaving by taking folks below 100, not by shifting the odds - and it feels like the cutoff does move significantly season over season. more than it should if the game was giving the higher odds to the next man up, so to speak? no idea.

hey benis, buddy, could you figure that one out for us?
Yeah I'm not sure exactly how it decides who leaves early. My guess was:

Each position on the big board has a defined chance of leaving early (98% for players ranked 1-5 for example). The sim runs through each player and decides whether they should leave or stay. It goes through the entire top 100 and if the # of players needed is satisfied, it stops. If not, it runs through the list again. I'm not sure it really works this way since it doesn't account for players not on the board but I guess it's possible. I think this is too simplistic for it to really work this way but just spit balling.

As for the shifting odds theory - I'm not totally sure but I don't really think so. Based upon what I've seen, there is VERY little difference between a Likely Staying Sophomore ranked at 40 as at 80. The odds of those two players leaving is nearly identical. If the odds were shifted, you'd see a much more clear trend towards players ranked higher (at 40 vs 80) since as players in front of them decide to not leave early, their own ranking would increase into the 'very likely to leave' range (or at least it would see a significant bump in %).
1/3/2020 12:50 PM
I preface this by saying this is all just speculation on my part. I don’t care enough about the details to collect any data on it.

“Based upon what I've seen, there is VERY little difference between a Likely Staying Sophomore ranked at 40 as at 80. The odds of those two players leaving is nearly identical. If the odds were shifted, you'd see a much more clear trend towards players ranked higher (at 40 vs 80) since as players in front of them decide to not leave early, their own ranking would increase into the 'very likely to leave' range (or at least it would see a significant bump in %).”

40 isn’t the cusp of anything, though. Sophs at 40 are never going to jump into “likely going” and would very rarely, if ever, jump to “on the fence”, because the odds of 12 non-graduating guys in the top 40 staying are very, very slim. If there is a difference between the odds of the likely staying sophs at **30** and 80 (or 60, or 100, or whatever) is where you’d want to look.
1/3/2020 1:25 PM
“Each position on the big board has a defined chance of leaving early (98% for players ranked 1-5 for example). The sim runs through each player and decides whether they should leave or stay. It goes through the entire top 100 and if the # of players needed is satisfied, it stops the big board, in order, until all draft slots are filled. If not, it runs through the list again the big board doesn’t fill the draft slots, it will start evaluating and considering the top guys not listed on the big board.
1/3/2020 1:46 PM
◂ Prev 123 Next ▸

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2026 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.