Can Anyone Answer An Age-Old Draft Pool Question?

[Chappy]

I was reading this thread, trying to find answers as to why the NPBL has had so few big prospects in each of the drafts since I've been commish (interestingly enough, the prior NPBL commish had also posted in that thread).

Another NPBL GM had proposed a theory much like the one Sporr throws out:

Originally Posted by sporr

I have always wondered how the game decides the talent to make available. Does it look at the talent already in the league (in which a league with tons of talent is balanced by having less talent in the coming drafts)? Or does it try to provide a certain level of talent in each draft, give or take?

Does anyone know a definitive answer to this question? Is it random?

When the NPBL was formed, each owner created his own stud player. Is it possible that these 24 superstuds have impacted the draft pools?

[Loompa17]

Now that I think about this, I think a few case studies should be able to answer this question. Keep all the league settings (financials, #of teams, etc., etc.) the same. The only variable to change is the ratings of the players.

Group 1 - Edit the player ratings so that there is a lot of talent already in the league.

Group 2 - Keep the player ratings at a "medium" level.

Group 3 - Edit the player ratings so that there is not much talent in the league.

For each group then, proceed to the next season and see if the talent available in the draft is similar or not.

Okay, it's easy to suggest the study, but carrying it out takes the time. Is there an easy way to edit a bulk of player ratings without going player-by-player in game? (Forgive me if this is a naive question, I've never tried to do something like this before).

[thomamon]

Maybe using the CSV's that OOTP Exports would be the best way to edit the ratings quickly then importing them again.

This is a great question. In one league I am in, they edit the draft every year and have way too many good players. I think it truly effects the league as we have hitters who have 7 & 8's in contact and bat .250 or lower consistently.

[TheRockCT]

I've always liked leagues that "for the most part" let the game produce whatever type of draft it wants to, but if there is some heavy skewing going on, then it's time to slip things back the other direction. If the season before there were 4 guys over 50 homeruns and an "ace" 5-star SP is rated with 6's or 7's across the ratings board, it's time to modify the SP's in the draft, and vice versa.

[disposableheros]

i did some studies on the draft aspect for the WHBL to decide how many rounds are ideal and although i didnt track what you are asking specifically, i found the draft is the same as the leagues stat output. there is a progression and flow to it. its hitter heavy for a while and moves toward the pitching side. then its pitcher heavy for a while and swings back to the hitters. its just like the eras in MLB. some years there are no high quality players to draft and some years there are an abundance. i think it all evens out over the long-term.

[Loompa17]

As we (Chappy really) did some research on the previous threads on this topic, it seems that the conclusion many have come to is just as you say:

Originally Posted by disposableheros

i think it all evens out over the long-term.

However, I don't think that necessarily answers the question as sporr presented. I can see how either option will, over time, even out.

[disposableheros]

i think one of the problems is that most leagues dont make it to "long-term" and if the league is lucky (is it luck?) enough to make it to "long-term", most of the owners dont. they come and go every handful of seasons and never get to see the other side of the coin.

it even happens to solo leagues (non-long-term)....they just dont last long enough to get through the cycle.

[Chappy]

Yeah, at this point, I'm happy to just sit back and see what happens over the next decade. Of course, 3 seasons from now we may (crosses fingers) be into OOTP7, meaning everything could change...

But I still wonder if the game looks at the current league talent when it generates the draft pool. My guess is no...

[Melo]

Originally Posted by Chappy

Yeah, at this point, I'm happy to just sit back and see what happens over the next decade. Of course, 3 seasons from now we may (crosses fingers) be into OOTP7, meaning everything could change...

But I still wonder if the game looks at the current league talent when it generates the draft pool. My guess is no...

I agree. I think it would be pretty difficult and time consuming to write the code evaluate the league and form the draft to keep it balanced. I think it's random and sometimes you will see an incredibly strong or weak draft strictly by chance.

What I like to do in my solo fictional leagues is edit the ages of the guys in the draft class. All the 23 & 24 year olds become 17 & 18 year olds. What this does is broaden the pool of players who may raise themselves to the next level.

[YoungEagle]

Chappy,

I think I have a pretty easy way to test the draft pool. Once the National Cup (the championship game in the NPBL) has finished, exit the game before proceeding to the next season. Using explorer, make a couple of copies of the current league file. Open each one individually and proceed to the next season, then compare the draft pools from each. If there is variability in the talent levels between each draft, then the game doesn't take the current league talent levels into consideration when forming the draft pool.

William
former NPBL owner: Colorado

[Loompa17]

Originally Posted by YoungEagle

If there is variability in the talent levels between each draft, then the game doesn't take the current league talent levels into consideration when forming the draft pool.

But if there is not significant variability in the talent levels, then nothing has been proven.

[Loompa17]

I still think the best way to test this is as I had described in post #2

I'm getting towards the end of a season in my solo league. When I get there, I'll try this experiment and see how it goes.

[ctorg]

From my experience, the game simply generates random players according to the modifiers. What this means, of course, is that the number of draft rounds greatly affects the quality of players available in the draft. If you have one draft round, you'll probably only have one or two really good players in the whole draft. If you have 25, you'll have a lot more, and so on as you increase rounds.

This effectively means that different leagues will have different average talent levels for starting players, since leagues with more draft rounds will have more higher-rated players. This doesn't sit right with me, and I wish it was different, but from what I can tell, that's how it works.

[Loompa17]

I'll bet your right ctorg. In running my "experiment", I may try using # of draft rounds as another variable to change (after I try changing the existing talent variable).

[Chappy]

I'm in both the JOBL (16 teams, 5 rounds, pool of 80 players) and the NPBL (24 teams, 4 rounds, pool of 96 players).

There always seems to be more talent available in the JOBL drafts than the NPBL drafts despite the fact that the JOBL has fewer total players in the pools. I don't know for sure, but I assume Jason uses the default settings. This is strange to me... but it could be the luck of the draw.

[gubaru]

Originally Posted by Chappy

I'm in both the JOBL (16 teams, 5 rounds, pool of 80 players) and the NPBL (24 teams, 4 rounds, pool of 96 players).

There always seems to be more talent available in the JOBL drafts than the NPBL drafts despite the fact that the JOBL has fewer total players in the pools. I don't know for sure, but I assume Jason uses the default settings. This is strange to me... but it could be the luck of the draw.

IME, the number of rounds matters more than the number of players.

[Vris]

Have you tried generating a bunch of players, but with less rounds? ie, under league setup, have it generate 10-15 rounds of players, but in the pre-draft screen, cut it back down to 4 rounds like you prefer. More players are generated, and you have a better chance of getting those top-notch prospects.

I think this, as well as the creation modifiers have a lot to do with the draft class. I've never put together a comprehensive study, but in one of my solo leagues, I upped the modifiers (hitting and pitching) and the classes after that were largely extremely biased towards incredible hitting. I knocked it down quite a bit since then.

[Chappy]

Originally Posted by gubaru

IME, the number of rounds matters more than the number of players.

Hmmmm. Interesting. That would explain the differences in the JOBL and NPBL draft pools.

[Loompa17]

Originally Posted by Loompa17

Keep all the league settings (financials, #of teams, etc., etc.) the same. The only variable to change is the ratings of the players.

Group 1 - Edit the player ratings so that there is a lot of talent already in the league.

Group 2 - Keep the player ratings at a "medium" level.

Group 3 - Edit the player ratings so that there is not much talent in the league.

For each group then, proceed to the next season and see if the talent available in the draft is similar or not.

I ran this experiment, pretty much as I described above. The league has 24 teams and a 4 round draft, scouts are disabled. If anybody's interested in other league settings, let me know.

For Group 1, I took about 100 2 star (some gold star, some blue star) players and edited their ratings so they were all 5 star. Group 2 I left the same. For Group 3, I took everybody who was 3.5 stars or better and edited them so they were all 1 star.

I made 4 copies of each group and then proceeded to the next season (which created the amateur draft players). Four isn't a great sample size, but at least it's better than 1.

For Group 1 (HIGH TALENT ALREADY IN LEAGUE), the players to be drafted broke down like this:

Stars       #1       #2      #3     #4
5            2       1       0       3
4            1       2       4       5
3           10      11      11       9
2           33      28      29       26
1           50      54      52       53

For Group 2 (NORMAL TALENT):

Stars       #1       #2      #3      #4
5            1       6       5       5
4            2       4       0       2
3            7      12       8       5
2           35      28      37       31
1           51      46      46       53

And Group 3 (LOW TALENT ALREADY IN LEAGUE):

Stars       #1       #2       #3     #4
5            4       4       2       1
4            6       4       0       2
3            5       8       6       8
2           29      33      29       40
1           52      47      59       45

For there being such a big difference in the amount of talent already in the league, there does not appear to be much of a difference between the talent levels of the amateurs. So I am thinking that this is not a factor that's used when creating new players for the amateur draft.

Next, I'm going to run some case studies on this league where there are 2,4,8 & 12 rounds. I'll post those results later.

[Chappy]

Yeah, it looks to be pretty random... I'd say those pools are similar to the ones we've experienced - where you're drafting 2-star players by the middle of the first round. I'm not disappointed in the number of 5 star prospects, but the 4-star guys are just as scarce - I think it should be more of a bell curve...