Prospects in Historical Leagues to reflect how they were viewed at the time

[Gammajamma]

I love the game and one of my favorite aspects is creating alternative histories. While OOTP is incredible in this, I do wish that the top prospects when you start the league was more accurate to what they were at the time. It feels like hindsight plays a little bit of a role to how they pan out/are rated which is natural. I noticed that a lot of the top prospects who flamed out are never highly regarded in my restarts at the beginning. Part of the appeal is having the alternative timeline is making guys stars and hall of fame talents when in reality they fizzled out. It feels cheap to me when I edit the guys to adjust their potential and I was wondering if anyone knew what I could do to have it be portrayed more accurately? For example Here are a few top 50 prospect lists from 2004-07

[https://www.mlb.com/news/2007-top-50...ist-c301612226)

[https://www.mlb.com/news/2006-top-50...ist-c301623478)

[https://www.mlb.com/news/2005-top-50...ist-c301622238)

[https://www.mlb.com/news/2004-top-50...ist-c301622276)

​

I want it so it reflects Brandon Wood, Delmon Young, BJ Upton and others as top level prospects with high potential. I want to try and win with the corey pattersons of the worlds. Any help or suggestions is much appreciated

[spencer10177]

Following, I would like to be able to do the same thing.

I would also like for some of these prospects to pan out unexpectedly.

[kriscolic]

I too would love to see (as a lifelong Mets fan) a universe where Fernando Martinez develops into a perennial MVP candidate.

(What follows is probably oversimplified, but in the interest of brevity...) As currently constructed the OOTP development engine is geared toward a singular outcome aka a player's "potential". In order for the game to produce universes wherein Todd Van Poppel and Andy Marte become stars a not-insignificant amount of the time while turning out the way they actually did in most cases, I believe the game would have to be coded to include multiple, distinct possible outcomes. Each player would have to have a range of possible development paths and particular odds of reaching each of them. This would be similar to how projection systems like ZIPS and PECOTA work. Van Popple, for example, would have a, say, 1% chance of becoming Tom Seaver, a 10% chance of becoming Jack Morris, a 40% chance of becoming, I dunno, Pete Harnisch, etc etc.

I have to think that implementing something like this would be a huge undertaking for the developers and is unlikely to happen any time soon. But here's hoping!

[Charlie Hough]

There are a few things you can do to help in this area.

1. You can create a historical game and choose the option to have player potentials based on the peak seasons of their careers rather than their entire career. This will theoretically allow them to develop to the best potential they realized in real life. It also excludes the negative effects on their potential from real life seasons where they later flamed out or rapidly declined in performance. Using peak seasons doesn't guarantee that these players will successfully develop to that level or sustain it. But it will get you closer to what you want to see in your games.

2. Turn off recalc and only use the development engine for all future player ratings and development. You will see many players fail to reach their potential or actually exceed it. Their potential from their original historical import will strongly influence their career arc and eventual results, but with the development engine there is still a lot of randomness, and their potentials can change dramatically. If you combine these two options, I suspect you'll be pretty happy with your results.

[HumanRainDelay]

Originally Posted by Charlie Hough

There are a few things you can do to help in this area.

1. You can create a historical game and choose the option to have player potentials based on the peak seasons of their careers rather than their entire career. This will theoretically allow them to develop to the best potential they realized in real life. It also excludes the negative effects on their potential from real life seasons where they later flamed out or rapidly declined in performance. Using peak seasons doesn't guarantee that these players will successfully develop to that level or sustain it. But it will get you closer to what you want to see in your games.

2. Turn off recalc and only use the development engine for all future player ratings and development. You will see many players fail to reach their potential or actually exceed it. Their potential from their original historical import will strongly influence their career arc and eventual results, but with the development engine there is still a lot of randomness, and their potentials can change dramatically. If you combine these two options, I suspect you'll be pretty happy with your results.

Interesting topic GammaJamma, and I appreciate the improvement suggestion for this, CharlieHough. With your latter suggestion regarding using the development engine to generate an outcome for a historical rookie's progression, I had a more general question for the community.

Say you are playing a 1980s historical league and allowing the development engine only to predict how players develop. In the second round of your amateur draft, you can select from several highly rated prospects including "Joe Nobody" and Bo Jackson.

As a general manager, should I completely ignore Bo Jackson and other familiar names and evaluate talent strictly using the numerous ratings?

I am guessing the correct answer is "yes" as the performance outcome is supposed to be random.

In practice though, I find myself surfing draft pools/free agent lists and recognizing players who had some success in their real life MLB careers--I am more attracted to these players as they have demonstrated a path to eventually helping my team as compared to Joe Nobody that I equate with more of a dice roll.

What do you think?

[David Watts]

Originally Posted by Charlie Hough

There are a few things you can do to help in this area.

1. You can create a historical game and choose the option to have player potentials based on the peak seasons of their careers rather than their entire career. This will theoretically allow them to develop to the best potential they realized in real life. It also excludes the negative effects on their potential from real life seasons where they later flamed out or rapidly declined in performance. Using peak seasons doesn't guarantee that these players will successfully develop to that level or sustain it. But it will get you closer to what you want to see in your games.

2. Turn off recalc and only use the development engine for all future player ratings and development. You will see many players fail to reach their potential or actually exceed it. Their potential from their original historical import will strongly influence their career arc and eventual results, but with the development engine there is still a lot of randomness, and their potentials can change dramatically. If you combine these two options, I suspect you'll be pretty happy with your results.

I wonder if along with your suggestions, one should change Talent Change Randomness from the default 100 to 200? Just curious?

[Charlie Hough]

Originally Posted by HumanRainDelay

Say you are playing a 1980s historical league and allowing the development engine only to predict how players develop. In the second round of your amateur draft, you can select from several highly rated prospects including "Joe Nobody" and Bo Jackson.

As a general manager, should I completely ignore Bo Jackson and other familiar names and evaluate talent strictly using the numerous ratings?

If you are not using recalc at all and you're strictly relying on the development engine, then evaluate players based on their current and potential ratings, just as you would for fictional players. If your game is importing players and calculating the current ratings and potential based on their real life peak seasons, then Bo Jackson is probably going to be a great prospect. But if there's somebody who looks considerably better now and in the future, then don't let the name determine your decision.

Of course, it can be fun to go ahead and draft Bo Jackson since he's a name you know from real life. If he becomes as good as he was in real life and has a longer career, maybe you'll enjoy having him around all the more. But if Joe Nobody ends up being a huge star for another team, you might end up kicking yourself.

[Charlie Hough]

Originally Posted by David Watts

I wonder if along with your suggestions, one should change Talent Change Randomness from the default 100 to 200? Just curious?

I have never modified TCR, so I can't provide a recommendation there. One thing to keep in mind is that changing this setting makes it more or less likely that players will experience significant changes in talent over the course of their career. I take this to mean that there will be more random variations compared to a player's current and potential ratings, so players might be a lot less "stable" or "reliable" over the arc of their careers. Also, I've seen discussion elsewhere that TCR doesn't have as much of an effect on older players. It has most of its effect on guys in their late teens to late twenties.

Here's what the OOTP manual says about it:

The Talent Change Randomness option is a numeric value from 1 to 200 that controls how random player talent changes are. For example, a 200 here would mean that talent changes are highly random, making it more likely that players would experience significant changes in talent over the course of their career. Tweak this if necessary if you feel that player talent changes are either too drastic or too conservative. 100 is the default.