Serious Problems with Feeder Leagues?

[MidKnight]

Get ready for a long one...

Well, after finally getting my league in place, what do I do? Try to add more to it! I read a couple threads about the feeder leagues and figured I'd give it a go. My current league set-up is a 16 teams ML, with all levels of minors (AAA, AA, A, SA, R), as well as two international leagues (12 teams each). It's the "Fictional Universe in a Box" template package I released in the mods sections. I saw varying formulas for the number of teams needed to have, but they ended up with similar numbers, so I stuck with the formula in the manual.

I've heard of some issues of too much talent ending up in the draft pool, so I decided to investigate. I came up some serious issues and ran a series of tests which took about 3 hours or so. I'm hoping someone can either point out what's going wrong, or if it's nothing I'm doing, that it'll be addressed. In the current state, I've found that feeder leagues only destroy leagues - - either by not creating enough players or by creating too much talent. I'll explain...

Info: My Major League had defaulted to 20 draft rounds/21 generated. Upon adding feeder leagues it jumped to 25/26. The feeders have an age range of 14-18 for High School (HS) and 18-22 for college (NCAA). High school has a 45 game schedule, college 58. I tried both having the ML only allow the feeder leagues in the draft AND have it also generate additional players. I looked at talent by counting the potential ratings of each draft pool - - I looked at 5, 4.5, 4, 3.5, 3, and 2.5 stars and counted how many of each appeared in the draft pool (scouts off), as well as how many total draftees were available.

I ran two sims without feeder leagues and came up with the following for star ratings:

4, 5, 6, 1, 2, 6 [84.5]
3, 4, 3, 5, 2, 2 [77]

For the first line that means there were four "5-star" prospects, five "4.5 star prospects"...etc. The number in the brackets was a rudimentary overall draft ranking with these numbers - I assigned a point value based on how many stars there were. For example, one "5-star" prospect was worth five points, one "4.5 star" was worth four and a half points. I added the totals together.

Being a 20 round draft, there were approximately 14 pages worth of draftees (if I recall, just about 336 draftees). Now for the feeder leagues. I used the formula in the manual (number of draftees needed, divided by six). That's [16 teams] * [25 rounds], then divided by six. That's roughly 67 teams. I always set it up so there were more NCAA teams than HS - - I wanted most players going through college.

In any case, for the first tests, I left the setting to "Use Feeders + Additional Players" for the amateur draft, just in case there weren't quite enough to fill all 25 (no longer 21) rounds. These are the results:

60 Teams - 32 NCAA, 28 HS
5, 14, 12, 5, 5, 12 [198.5]

That was waaaay too much talent (more than twice a normal draft), so I decided to decrease the number of teams.

44 Teams - 32 NCAA, 12 HS
13, 12, 13, 10, 10, 4 [246]

That didn't work. Figured it must have been a fluke. I lowered the teams again.

36 Teams - 24 NCAA, 12 HS
15, 19, 7, 10, 3, 9 [255]

12 Teams - 8 NCAA, 4 HS
3, 11, 7, 9, 5 [193.5]

Even a paltry 12 teams generated too much talent. Then it hit me, I'll turn off the "Generate Additional Players" and set it to "Feeder Leagues Only". Well, the first result with 60 teams came out much better talent-wise:

60 Teams - 32 NCAA, 28 HS - Feeder Leagues ONLY
1, 2, 4, 3, 2, 1 [49]

Results were on the low end, but that could be fixed by adding more teams, I suspect. The only problem was...there were only about 90 draftees in the pool - - not nearly enough for a 25-round draft which requires about 400. I heard many saying that with a 5 year age range (14-18, 18-22), about 6 draftees will leave a team each year. Instead, it only appeared to be about 1 or 2 were leaving each team. It took me until I had 190 teams before I even had enough players for a 20-round draft (about 340).

In summary...what's going on? I figure I'm either doing something horribly wrong, or feeder leagues aren't functioning properly. My methods of measuring draft talent are rudimentary at best, but they should be decent enough for gathering a general look at potential. If I set the amateur draft to "Feeder Leagues + Additional Players", there is far too much talent in the league. If I set it to only use feeder leagues, it won't generate enough players. In any case, let me know...I spent a good three or four hours on this last night, and I'm hoping I'm just missing some minor detail.

[Charley575]

Watch out. I have found that in the INITIAL YEAR, draft size is low, but in subsequent years, it is in fact about 6 a team. I suspect this has to do with an age range of, for example 18-21, which means that players will "retire" to the MLB draft pool at 22. At initial creation, the game will generate players between 18 and 21, it will not generate any 22 year olds. Therefore only those players who are 21 years old AND have a birthday before the amateur draft will enter the draft pool in the first year.

[MidKnight]

Originally Posted by Charley575

Watch out. I have found that in the INITIAL YEAR, draft size is low, but in subsequent years, it is in fact about 6 a team. I suspect this has to do with an age range of, for example 18-21, which means that players will "retire" to the MLB draft pool at 22. At initial creation, the game will generate players between 18 and 21, it will not generate any 22 year olds. Therefore only those players who are 21 years old AND have a birthday before the amateur draft will enter the draft pool in the first year.

Thanks for the heads up on that. I just simmed a bit and you're correct - - the draft pool with a 60 team setup was 212 the first year and then the proper 412 the next. The only problem is...there's waaaayy too much talent again. I ended up with 20 "5-star prospects" alone! The first year was real good, too (about 10 5-star), but it just seemed to get worse the next year. Battists, or anyone else...any clues?

[MidKnight]

dola,

I'm seeing another problem. In the initial year, a lot of young HS players are being released for no apparent reason. I'm talking 14 year old, 5-star prospects. Because they were released, they appeared in the draft pool - - so you have a bunch of 16 year old superstars in minor league organizations a couple years later. I haven't seen this occur after the initial year, but I could be wrong.

[NomarHits400]

Originally Posted by MidKnight

dola,

I'm seeing another problem. In the initial year, a lot of young HS players are being released for no apparent reason. I'm talking 14 year old, 5-star prospects. Because they were released, they appeared in the draft pool - - so you have a bunch of 16 year old superstars in minor league organizations a couple years later. I haven't seen this occur after the initial year, but I could be wrong.

Midknight

This is a problem- and the PCM's MUST be modified...I will post mine for a 124 team NCAA feeder and 250 HS feeder if you like- though it sounds like you are doing less than that- but it can be done.

Also, the poster who said hte first year yield is low is correct- I am getting only about 12-15 4 star and up players the first year- but it goes up and steadies to around 20-25 overall throughout the years. I am looking 11-12 years into the sim now to see how the talent distribution is comparing to a real 2007 MLB league- and it still needs a little work- but looks good overall...

[MidKnight]

Originally Posted by NomarHits400

Midknight

This is a problem- and the PCM's MUST be modified...I will post mine for a 124 team NCAA feeder and 250 HS feeder if you like- though it sounds like you are doing less than that- but it can be done.

Also, the poster who said hte first year yield is low is correct- I am getting only about 12-15 4 star and up players the first year- but it goes up and steadies to around 20-25 overall throughout the years. I am looking 11-12 years into the sim now to see how the talent distribution is comparing to a real 2007 MLB league- and it still needs a little work- but looks good overall...

Agreed. I simmed a couple years in the future - and lo and behold - I'm still seeing young guys being released by their High School OR College team and entering the draft early. I've yet to find a solution, and this is a major problem.

Ahh, so the Sabermetic Player Modifiers should adjust the potential? I'll have to take a look into it. I'd be thankful if you could post your PCM's so I have a bit of reference.

[pappyzan]

this probably should be addressed in the technical support forum, maybe a TT# would have a higher chance of getting assigned to these feeder league issues.

here are my quick, unscientific findings concerning feeder leagues:
http://ootpdevelopments.com/board/sh...27#post2130827

i really like the feeder leagues option, it appears it just needs a bit of tweaking from Markus before it's potential can be fully utilized within the game. just waaaay too many good-to-great draftees no matter how many rounds are used, no matter how many feeder leagues or feeder league teams are used, and no matter how big/small your league/universe is.

it also appears to generate too many bad-to-awful draftees when feeder leagues are used in any way within the game. really needs a good all-around tweaking to smooth out the talent generated from the feeder leagues.

[MidKnight]

Good call; I'm going to make a Tech Support thread pointing to this one.

[professor ape]

Originally Posted by NomarHits400

Midknight

This is a problem- and the PCM's MUST be modified...I will post mine for a 124 team NCAA feeder and 250 HS feeder if you like- though it sounds like you are doing less than that- but it can be done.

Also, the poster who said hte first year yield is low is correct- I am getting only about 12-15 4 star and up players the first year- but it goes up and steadies to around 20-25 overall throughout the years. I am looking 11-12 years into the sim now to see how the talent distribution is comparing to a real 2007 MLB league- and it still needs a little work- but looks good overall...

Please post your PCMs

[rudel.dietrich]

I too had to turn off feeder leagues
They simply dump too many good players into the league and it kills your league after a few seasons and makes the draft meaningless

Here is another problem im having
You can set the creation ages for the feeder leagues. Lets use a HS league for an example.
I set these as the min and max creation ages as 14
That is good
The problem is that upon creation ALL players are 14.
Well you say, why dont you just set it to 14-18?
Well that solves the initial creation problem but for every near year of creation, it puts in players between 14-18 years old which is not what I want!

There needs to be two options. Age range for initial creation, and age range for normal creation.
I think im going to go over and TT this right now

[MidKnight]

A shameless bump to draw attention to the issue or get some good PCM's from the community.

[NomarHits400]

Originally Posted by MidKnight

A shameless bump to draw attention to the issue or get some good PCM's from the community.

I apologize MidKinight- I meant to get these to you yesterday- but I wanted to run a few more tests to make sure things are stable 10-20 years out...I will post them shortly.....

[MidKnight]

Oh, no worries...didn't mean to rush. Just wanted to keep the thread off the second page for a bit.

[NomarHits400]

Originally Posted by MidKnight

Agreed. I simmed a couple years in the future - and lo and behold - I'm still seeing young guys being released by their High School OR College team and entering the draft early. I've yet to find a solution, and this is a major problem.

Ahh, so the Sabermetic Player Modifiers should adjust the potential? I'll have to take a look into it. I'd be thankful if you could post your PCM's so I have a bit of reference.

PCMS....

Ok, a little background on my league first- just for reference- because it will probably have some effect if you use a different league size/feeder league size than mine- but I think these are WAY better than the defaults:

30 Team MLB with full minors- I happen to be using a versioin of Cubby's rosters- so minors are partially stocked already with a decent ratio of talent.

I have a 112 team NCAA - David's excellent template that uses real NCAA powerhouses.

I also have a 2 sub league- 8 divisions each- 15 teams per division -240 team total HS feeder league.

For the NCAA feeder- important notes:

The sabermetric player creation modifiers for this league are ALL set to 0.83. Lowering them will lower the initial player creation ratings for all players- but it will NOT effect their potential- raising it will do the opposite. Why did I decide on 0.83? It will allow for a select few of the best colleger players to be 2.5 and 3 star players- or ready for the bigs almost right out of college- ala Chad Cordero and Huston Street- maybe 3-5 of these per draft. It also is high enough to distinguish the head start college players have over high school players (read: less minor league development). I have no teenagers in the MLB after 10 years- and only about 10-15 players 20-22...So lower this if you want fewer guys with MLB type ratings out of school- and to decrease overall ratings- increase these for the reverse effect.

For the HS feeder:

Same idea as above- but all sabermetric player creation modifiers for this league are set to 0.60. Conversly, I don't want ANY high school players ready to step into the bigs- and with 0.6 you won't get any- it will also take the average 19 year old HS player a 1-3 years longer to develop- but again- you can modify in the above fashion for college guys to suit your taste.

For your MLB parent league:

I am running a 25 round draft- feeder leagues only (there are plenty of players to feed a 30 team MLB).

Sabermetric player creation modifiers were not specific enough for me to get the results I wanted for the Majors- so I have them all at their default settings of 1.000. Easy enough.

However, I have modified ALL Traditional OOTP Player Creation Modifiers:

Batting Contact 0.88
Batting Gap Power 0.75
Batting Power 0.80
Batting Eye 0.75
Batting Avoid K's 0.83
Piching Stuff 0.90
Pitching Movement 0.86
Pitching Control 0.86
Running Speed 0.78
Fielding Ratings 0.81


I have messed with these a lot- and in general these are the effects I am seeing:

The talent will be a little thin for a few years- but it gets better, and about ten years in the top 100 prospects has good balance- and they are real talent comparable to the MLB group I started with in 2007.

A couple of annoying things: I have power tuned down a bit- and it may seem that there are not a whole lot of great hitting prospects power and average wise early on. But in about 5-7 years you will see a dozen or so prospects with good chances of being 80-90 in the power department- even though the draft classes leading up to these years all were maxed out in the mid 70's at the highest...You can obviously change these if you like- but beware of consequences down the road. This is because quite a few guys that were not initially highly rated are developing into good players after the draft.

Another thing about these feeeder league settings- that you may like or hate- the draft is a real crapshoot now. High draft picks on the whole are better- but their are literally dozens of great players in the later rounds- that develop out of nowhere- I kind of like this- but you may not if your #1 overall selection flops and is out of the bigs in 4 years (it happened 4 times in ten years in one sim). If you want to add a little more predictability to your universe- go to "Player and Picture Options" in your league setup and turn talent change randomness down to 50 or so- and this should decrease the chance that your first round stud will flop....To each his own.

Another thing I have noticed- that really has nothing to do with these modifications- but I am putting it out there in case anyone can figure out how to fix it: The AI is dropping SCADS of first round picks in the September after they were drafted- many after posting good seasons...And I am not sure why. It may be related to the roster size limits I have imposed (25 for all but the Rookie level teams- which I have at 35)- but I am not sure...This is really a bummer for me- but I am going to continue to look at it.

[Underlined.i]

Have you ever looked at the players actual potential ratings instead of just potential stars? I see a lot of MR that have potential ratings of something like 5-6-6 and still have a 5 star potential. Maybe the abundance of 5 star potential doesn't matter if they really aren't that good.

[NomarHits400]

Originally Posted by Underlined.i

Have you ever looked at the players actual potential ratings instead of just potential stars? I see a lot of MR that have potential ratings of something like 5-6-6 and still have a 5 star potential. Maybe the abundance of 5 star potential doesn't matter if they really aren't that good.

That's a good point- but I look at the ratings for potential themselves as well as the stars.

Also, I will note that using the set-up I have above I see VERY few highly rated MR's each draft- maybe 3 or 4 4 star and up total in the whole draft....Lots of closers though.

[rudel.dietrich]

I did some messing around with PCMs and had mixed success
And it really should not take that much work to get the leagues set up and working in a somewhat realistic manner.
So sadly unless some major changes take place I am no longer going to use feeder leagues

That was my most looked forward too feature.
But beyond that, im still very happy with the game overall.
Its just a little annoying that no beta tester seemed to bring up this issue and I dont like when featuers are toted before release and then come to find out they work half right.
And I know someone will quote me and try and tell me they do work, and they do work to a degree.
But far too much work and testing is needed to get them working where the first two rounds of your draft are not 5 star players.

After a few years and every team is filled with 5 star players, the stats balence out. But it just looks bad and as I said before, it makes the draft meaningless since no matter if your finish 1st or 20th your going to get be able to draft a 5 star player.

[Underlined.i]

Originally Posted by Fallschirmjager

After a few years and every team is filled with 5 star players, the stats balence out. But it just looks bad and as I said before, it makes the draft meaningless since no matter if your finish 1st or 20th your going to get be able to draft a 5 star player.

Does this happen with the PCMs? Forgive my somewhat ignorance as I just got the game a few days ago. I was an avid 6.5 played but passed on 2006. The feeder leagues are definetly an awesome feature if they didn't create a super draft. I have heard a lot of complaints that the star feature is off in this game as well...so when your league is being overrun by 5 star players are they actually rated like a 5 star player should be? Or might they just have 6s and 7s across the board with a skewed 5 star rating?

[rudel.dietrich]

Originally Posted by Underlined.i

Does this happen with the PCMs? Forgive my somewhat ignorance as I just got the game a few days ago. I was an avid 6.5 played but passed on 2006. The feeder leagues are definetly an awesome feature if they didn't create a super draft. I have heard a lot of complaints that the star feature is off in this game as well...so when your league is being overrun by 5 star players are they actually rated like a 5 star player should be? Or might they just have 6s and 7s across the board with a skewed 5 star rating?

The star rating is in relation to every other played in the universe.

So if a player has 3s across the board but every other player has 1s then they will be a 5 star player since realative to the league they are a god like player.

But the ratings are pretty much 7-10 in mixes that make the players 5 star.
So yes, they are what you would think 5 star players to be.

In a 16 team ML the first 2.5 rounds are 5 and 4 star players when you do not mess with the PCMs beyond what they defaults are.
That is with a 40 team feeder league (20 team college and 20 team HS)

That is with a 15 round draft (to cover the ML and 2 minor leagues)
Doing the math set out in the guide 40 teams is right around the sweet spot you want.

But the problem is that each feeder team has 3-4 five star players at any given time.
1-2 of these might enter the draft in any given season.
Multiply this by 40 and add in the 4-3 star players and you have a serious problem.

In entire HS and college conferences are lucky if they produce one '5 star' player

The problem just needs some attention given to it and the talent level present in feeder leagues needs to be auto adjusted based on universe size.
I am no coder but I think the coding itself would be pretty easy since the draft already does such a thing.
It just needs some dedicated people to test out the leagues and give Markus the data on about where it should be.

The point that feeder leagues need to reach is where they export the same amount of talent as a draft would be if you let the game create all the players 'on the fly' i.e. not using feeder leagues.

So while not broken, feeder leagues require waaaaay too much effort to get to work properly.

As to your question, if you have enough time and patience, you can finally get the PCMs correct and find something your happy with.
But it takes alot of minor tweaking and then you have to sim out 6-8 seasons to let the league unfold.
That take awhile
Multiply that by several dozen depending on picky you are and how lucky you are and you will begin to understand how long it takes.
And it has to be adjusted for every ML size and every level of minors you use.
Just going from a 16 team ML to a 18 team ML will throw off the PCMs.
There is no golden set of PCMs that I have found to work on all league sizes.

So...they work, but it really shouldn't take this much effort to get a game feature to work 'correctly'
And IMO this is the biggest flaw with the game right now and should be priority #1 in any future patches.

Others may disagree or think that im wrong, but thats ok.
I have been wrong alot before.
But I have put alot of time into this issue and think I have a pretty good idea of where it stands and where it needs to be to work correctly out of the box with a minimal amount of set up time.