Are there any improvements to the feeder system?

[Cryomaniac]

Originally Posted by lukasberger

They also do have stats generated, though not usually four years. I wouldn't make sense to generate four years typically since most college guys are drafted and signed as juniors, or draft eligible sophs or juco players, not seniors.

Most HS guys also don't have four years of history irl, as they often don't play, or play super sparingly as freshmen

Would it be possible to work out roughly what percentage of drafted players have played for 1, 2, 3 and 4 years (at either college or high school) and then have something like that in the game? So say if 60%* of players are drafted after their junior year, then 60% of players would have 3 years worth of stats?

*I'm pulling numbers out of my backside, but hopefully you get what I mean.

[Lukas Berger]

Originally Posted by Cryomaniac

Would it be possible to work out roughly what percentage of drafted players have played for 1, 2, 3 and 4 years (at either college or high school) and then have something like that in the game? So say if 60%* of players are drafted after their junior year, then 60% of players would have 3 years worth of stats?

*I'm pulling numbers out of my backside, but hopefully you get what I mean.

Yeah, I get it. I imagine it's possible.

I'm not gonna do it though, as it's a ton of research and I just don't care that much, and I doubt Markus wants to.

If someone does the work and sends me the data, I can talk to Markus about getting the generated stats to reflect that.

[mpejkrm]

One more note - I'd like to see the idea of players that don't sign a little more realistic... getting them three years in college rather than one. And I wouldn't mind seeing a "commitment" on high school players so we can see where they would end up if they don't sign.

[Lukas Berger]

Originally Posted by mpejkrm

One more note - I'd like to see the idea of players that don't sign a little more realistic... getting them three years in college rather than one. And I wouldn't mind seeing a "commitment" on high school players so we can see where they would end up if they don't sign.

Yup. Working on getting the first to happen.

The second would be cool, but I think it's likely to take a bit longer.

[jpeters1734]

Originally Posted by lukasberger

Yeah, I get it. I imagine it's possible.

I'm not gonna do it though, as it's a ton of research and I just don't care that much, and I doubt Markus wants to.

If someone does the work and sends me the data, I can talk to Markus about getting the generated stats to reflect that.

I don't think a big research project is necessary. I think most people would rather see 4 years of stats over just one. Then if someone really cared enough to see a proper percentage of players that played 3 or 4 years, then they could do the research then.

Come to think of it, doesn't the game already do this when it generates college players with various ages? A player at 22 would have 4 years of stats. 21 would have 3, and so on.

[jpeters1734]

Another suggestion that was lost in this thread was having a continuous amateur list rather then the game generating players right before the draft.

Each player would generate stats every year and eliminate the need for feeders. You could also keep tabs on that player you failed to sign, especially if they go to college for 3 years.

[Cryomaniac]

Originally Posted by jpeters1734

I don't think a big research project is necessary. I think most people would rather see 4 years of stats over just one. Then if someone really cared enough to see a proper percentage of players that played 3 or 4 years, then they could do the research then.

Come to think of it, doesn't the game already do this when it generates college players with various ages? A player at 22 would have 4 years of stats. 21 would have 3, and so on.

As it stands players only generally have 1 or at most 2 years of stats, but your idea makes a lot of sense.

[SirMichaelJordan]

Originally Posted by jpeters1734

Another suggestion that was lost in this thread was having a continuous amateur list rather then the game generating players right before the draft.



Each player would generate stats every year and eliminate the need for feeders. You could also keep tabs on that player you failed to sign, especially if they go to college for 3 years.

Or just have feeder leagues that run in background like EHM.

Pretty much the sane thing you are suggesting but with more immersion factor.

[Lukas Berger]

Originally Posted by jpeters1734

Come to think of it, doesn't the game already do this when it generates college players with various ages? A player at 22 would have 4 years of stats. 21 would have 3, and so on.

That's what I was thinking, which is why I was a bit confused by your suggestion initially.

Originally Posted by jpeters1734

Another suggestion that was lost in this thread was having a continuous amateur list rather then the game generating players right before the draft.

Each player would generate stats every year and eliminate the need for feeders. You could also keep tabs on that player you failed to sign, especially if they go to college for 3 years.

Love this idea, definitely going to suggest it.

[Lukas Berger]

Originally Posted by SirMichaelJordan

Or just have feeder leagues that run in background like EHM.

Pretty much the sane thing you are suggesting but with more immersion factor.

That's one of my dreams for OOTP

[SirMichaelJordan]

Originally Posted by lukasberger

That's one of my dreams for OOTP

Yea, I peak over in the FHM thread from time to time and it looks like they are inspired by this model as well. BTS seem to be doing the same so hope this is in OOTPB's future!

[jpeters1734]

Originally Posted by lukasberger

That's what I was thinking, which is why I was a bit confused by your suggestion initially.

You may have misinterpreted what I meant. but maybe not. If a player is created at age 22, he should have 4 years of stats. Currently he would only have 1 unless, he failed to sign the year prior.

If a player is created at 20, one can assume that he was a JUCO player and should have about 2 years of stats.

Another thought I had is that for game generated HS players that do not get drafted. Rather than making them all FA, how about having some, if not most, go to college and re-enter the draft later? That way could also eliminate feeders. You could set the game to create, say double the amount of rounds of the draft. The kids that don't get drafted would opt to go to college and would re-enter the draft as they would in real life. That would also give them year to year stats, for every year they don't get drafted at least.

[Lukas Berger]

Originally Posted by jpeters1734

You may have misinterpreted what I meant. but maybe not. If a player is created at age 22, he should have 4 years of stats. Currently he would only have 1 unless, he failed to sign the year prior.

If a player is created at 20, one can assume that he was a JUCO player and should have about 2 years of stats.

Yeah, makes sense.

Originally Posted by jpeters1734

Another thought I had is that for game generated HS players that do not get drafted. Rather than making them all FA, how about having some, if not most, go to college and re-enter the draft later? That way could also eliminate feeders. You could set the game to create, say double the amount of rounds of the draft. The kids that don't get drafted would opt to go to college and would re-enter the draft as they would in real life. That would also give them year to year stats, for every year they don't get drafted at least.

Right. I've talked about this (and a lot of the other stuff in this thread) with Markus at great length. It's on his agenda, though maybe not for OOTP16.

[jpeters1734]

Originally Posted by lukasberger

Yeah, makes sense.



Right. I've talked about this (and a lot of the other stuff in this thread) with Markus at great length. It's on his agenda, though maybe not for OOTP16.

thanks for having our back!

[darkcloud4579]

This is the first version I've invested in building a strong feeder infrastructure for my league and it's paid dividends as I get normal dudes with normal ratings and guys who are 1-star dudes still manage to be contributors on teams. You just have to set up your feeder league as the main source of talent for your league.

I even have it configured so most kids can get 3-4 of HS experience and even in some cases, undrafted guys I will assign to a college team and so...it's not automatic but it does work okay in general. Just a lot of heavy lifting to get it right.

[Honorable_Pawn]

Originally Posted by jpeters1734

The database prospects are not the issue here, at least for me. I completely understand why they wouldn't have histories. Game generated amateurs are what needs to be improved. I think most people would rather have the generated amateurs to have a more fleshed out history rather than to deal with feeders.

These are my thoughts, as well.

[Honorable_Pawn]

Originally Posted by darkcloud4579

Just a lot of heavy lifting to get it right.

Here is the main problem with the current design.


It really begins with that for how much work they are and for how much resources they use. The benefit that they provide is marginal, at best, when taken into context against the idea of game-generated stats.

1. With feeders the players will all be from the same 100 or so schools. It is such a big limitation. This causes the loss of immersion. This is a negative benefit of feeders.

2. The stats in feeders are terrible at predicting success. The individual "actual ratings" at the feeder level are so close to one another that the population mean is tightly clustered. The standard deviations a so small that might as well not exist. Most players are clustered around the mean so the stats at the feeder level are almost purely random.

What all of this means is that a lot of work was putting into setting them up and they use huge resources. And the overall benefit it probably a negative benefit.

[jpeters1734]

Originally Posted by Honorable_Pawn

Here is the main problem with the current design.


It really begins with that for how much work they are and for how much resources they use. The benefit that they provide is marginal, at best, when taken into context against the idea of game-generated stats.

1. With feeders the players will all be from the same 100 or so schools. It is such a big limitation. This causes the loss of immersion. This is a negative benefit of feeders.

2. The stats in feeders are terrible at predicting success. The individual "actual ratings" at the feeder level are so close to one another that the population mean is tightly clustered. The standard deviations a so small that might as well not exist. Most players are clustered around the mean so the stats at the feeder level are almost purely random.

What all of this means is that a lot of work was putting into setting them up and they use huge resources. And the overall benefit it probably a negative benefit.

I actually don't mind at all that players are from the same schools mostly. But I totally get why that would be an issue for some.

I can't say that I agree with the second point. I have always found that feeders, for the most part, are fairly good for predicting success. I haven't tried them in stats-only, though. I would like to hear a stats-only guy weigh in. Besides, I thought feeder stats went off potential more than current ratings.

[Mr. Marlin]

I've been playing stats only since OOTP 12 with feeders. I pay no attention to them aside from the stat leaders until the draft pool is created. While it is fair to say that players who dominate in feeder league play don't necessarily make it to the Majors they are much more likely to. I've found in my OOTP15 league that OPS & WHIP are the key stats for batters and pitchers respectively. The AI seems to focus on players that lead in those categories also.

But as much as I find them indispensable I'd love to not need to use feeders. For all the reason Markus and others have pointed out. The one thing I'm leery of is the stats on the game-created draft pool players. Are they supposed to be career stats or pre-draft season stats? Does anyone here know?

[Honorable_Pawn]

Originally Posted by jpeters1734

Besides, I thought feeder stats went off potential more than current ratings.

I'm not so sure about this one.

I know conversation has persisted on this topic yet I do not personally recall a conversation that settled the issue. If it is true that the stats for feeder leagues are based upon potential then my argument is diminished. Nevertheless, in its entirety, I still think a slimmer database is much more preferable to an anchor of "slightly better statistical history".

But are they really worth it?

I don't think so.