Most REALISTIC Settings for MLB

[RubeBaker]

I've been experimenting with the settings for the past couple of months & I have FINALLY figured them out. I'm currently manager/GM of the 2016 Chicago Cubs and I've came up with the following settings below:

AI Trading Frequency: Props here to OOTP user rjl518. I'm using the same settings he's employing for his 2016 Mets Replay. April: Very Low, May: Low, June: Average, July 1-15: High, July 16-31 (Trade Deadline): Very High, August: Low, September: Low

Stats & AI-League Totals: I selected the year 2015 from the drop-down menu and then clicked on "automatically calculate modifiers".

Player Rating Scales: I use 20-80, no stars. Overall rating based on AI evaluation, not pure ratings.

Injury Frequency: Again, props to OOTP user rjl518. Injury Frequency for April & May: Low, June & July: Normal, August & September: High.

Player Development Settings: Batter Aging Speed=.250, Batter Development Speed=1.000, Pitcher Aging Speed=.375, Pitcher Development Speed=.900, Talent Change Randomness=67

AI Trade Settings: Very Hard/Heavily Favor Prospects

Player Evaluation AI Settings: Ratings Weight=40%, Current Year Stats Weight=40%, Previous Year Stats Weight=10%, 2 Years Ago Stats Weight=10%. I also clicked on "recalculate GM tendencies based on these weights" and under Global Settings I clicked on "recalc all ratings + run full re-scouting".

Note: I did this on the first day when I started the replay.

Through May 9, 2016 on my replay the Chicago Cubs are 22-9, 5.5 games in first place. In real-life the 2016 Chicago Cubs were 24-6, 7.5 games in first place at this time.

My goal is to get caught up with the real-life team which is why I'm currently playing a combination of one-pitch mode and pitch-by-pitch mode. Once I catch up with the real-life Cubs, I'm going to manage strictly by pitch-by-pitch mode in real-time.

I can't say that these settings are for everyone, but so far they've worked for me and this is after doing a LOT of tinkering and experimenting.

[PSUColonel]

1. Default

2. Default

3. 2-8 (no ratings relative) nearly 70 percent of all MLB players are rated average in most tools/skills.

The 20-80 Scouting Scale and what the numbers all mean

4. High (realistic)

5. Default

6. 50/25/20/5

[Libid21]

I just made a thread about this a week ago, it's probably on the second page of this forum... But I dug up an old conversation that I had with Nutlaw about this and the settings I used to play on were...

Scouts on. Coaches on.
Ratings/Potential on the 2-8 scale. Stars on.
Trade on Hard, favoring neither veterans nor prospects.
Ghost players should be off. Dynamic league should be off.

What I'm interested in, however, is how you guys play "realistically". Do you allow the manager to handle the lineups and pitching rotations? Or do you override him and set those yourself? I would like to play strictly as the GM but a lot of the decisions that the AI manager makes in terms of line ups and depth charts are just dumb... The same question goes to the set strategies as well.

[PSUColonel]

stars should be off, scouts on, coaches on

I also use minor-league roster limits 25,25,25,30,35) and have never had issues as others have reported, but maybe I just haven't noticed them...but I never see highly ranked players in the free agent pool or on waivers.

[RubeBaker]

Originally Posted by Libid21

I just made a thread about this a week ago, it's probably on the second page of this forum... But I dug up an old conversation that I had with Nutlaw about this and the settings I used to play on were...

Scouts on. Coaches on.
Ratings/Potential on the 2-8 scale. Stars on.
Trade on Hard, favoring neither veterans nor prospects.
Ghost players should be off. Dynamic league should be off.

What I'm interested in, however, is how you guys play "realistically". Do you allow the manager to handle the lineups and pitching rotations? Or do you override him and set those yourself? I would like to play strictly as the GM but a lot of the decisions that the AI manager makes in terms of line ups and depth charts are just dumb... The same question goes to the set strategies as well.

I manage and use the same lineups Joe Maddon uses against a particular pitcher the Cubs faced in real-life 2016 unless it's someone they didn't face or one (or more) of my players is injured, someone's hot or cold, etc. And I also use real-life transactions (trades, releases, signings, etc.) Also, once I get caught up to where the real-life Cubs are I'm going to be managing and GM-ing the games in real-time.

[sprague]

Originally Posted by RubeBaker

I've been experimenting with the settings for the past couple of months & I have FINALLY figured them out. I'm currently manager/GM of the 2016 Chicago Cubs and I've came up with the following settings below:

AI Trading Frequency: Props here to OOTP user rjl518. I'm using the same settings he's employing for his 2016 Mets Replay. April: Very Low, May: Low, June: Average, July 1-15: High, July 16-31 (Trade Deadline): Very High, August: Low, September: Low

Stats & AI-League Totals: I selected the year 2015 from the drop-down menu and then clicked on "automatically calculate modifiers".

Player Rating Scales: I use 20-80, no stars. Overall rating based on AI evaluation, not pure ratings.

Injury Frequency: Again, props to OOTP user rjl518. Injury Frequency for April & May: Low, June & July: Normal, August & September: High.

Player Development Settings: Batter Aging Speed=.250, Batter Development Speed=1.000, Pitcher Aging Speed=.375, Pitcher Development Speed=.900, Talent Change Randomness=67

AI Trade Settings: Very Hard/Heavily Favor Prospects

Player Evaluation AI Settings: Ratings Weight=40%, Current Year Stats Weight=40%, Previous Year Stats Weight=10%, 2 Years Ago Stats Weight=10%. I also clicked on "recalculate GM tendencies based on these weights" and under Global Settings I clicked on "recalc all ratings + run full re-scouting".

Note: I did this on the first day when I started the replay.

Through May 9, 2016 on my replay the Chicago Cubs are 22-9, 5.5 games in first place. In real-life the 2016 Chicago Cubs were 24-6, 7.5 games in first place at this time.

My goal is to get caught up with the real-life team which is why I'm currently playing a combination of one-pitch mode and pitch-by-pitch mode. Once I catch up with the real-life Cubs, I'm going to manage strictly by pitch-by-pitch mode in real-time.

I can't say that these settings are for everyone, but so far they've worked for me and this is after doing a LOT of tinkering and experimenting.

Thanks for posting this.
I Had never thought of altering the trade settings and injury settings each month to reflect more how a season really goes. I am going to look more into this.
Cheers

[Spanish Lefty]

Very interesting changing the settings depending on the month, but I would forget jaja.

These are my settings:

- Scouting report updates: season start and end. (forces you to request a scouting report if it's trading deadline, for instance)
- Scouting accuracy: very low (varies a lot for prospects, not so much for veterans)
- Player actual ratings: none
- Player potential ratings 2 to 8 (be careful with youngsters)
- Other player ratings: 2 to 8
- Overall and potential rating: none displayed (makes a huge difference in difficulty)

- Injury frequency: high (realistic modern day)
- Batter and pitcher aging speed: .300
- Batter and pitcher dev speed: 1.350
- Talent Randomness: 170 (160-180)

- Trading difficulty: very hard
- Preference: Heavily favor prospects
- AI Settings: 25 / 50 / 20 / 5

Just for fun, every year, when the World Series ends, I go to the financial tab under "league settings" and multiply every cell by x1.03 to x1.05. Sometimes x1.02... others I leave it the same. Because, you know, inflation. I like to see how contracts are every year a bit higher, like in real life.

Concerning the aging and dev speeds, I wrote this in another post:

If you leave aging settings on default... Rookies will be about 24 and will go as low as 22. We can see that is not true nowadays, with players being promoted younger. With "1.300 development speed" you get a bunch with 22-23, a few with 21 and one 20-year player every now and then (Correa, Osuna, Rougned Odor, Bryce... types).

With "aging speed" you have to be much more aggresive. If you leave it with the default setting, David Ortiz and A-Rod are pieces of meat immediately. After running several 30-year simulations, .310 was the most fair approximation (always with high frequency injuries). You get many real situations:

- Average player, durable or iron man, getting to 38-40
- Average player, fragile or wrecked, getting to 32 years, then struggling through the minors with some ML appearances
- Future hall of famers when 30, struggling with injuries, retiring at 35 and not becoming HOFers.
- Very good players, durable or ironman getting to 42-43... (rare, but possible)
... etc.

In the same 30-year simulations, the player which retired the oldest was Trout with 44. Most of the players still retire at 36-38.

The game becomes more real. With the default setting and high injuries everybody retires at age 38-39 MAX and there are no +4 WAR players over 32-33 years.

I just ran another simulation with .310 aging speed and 1.300 dev speed until 2027

Oldest players in ML (and probable similar players)

- 42: CL David Robertson (closing, Mariano Rivera) and RP Tony Watson (above average, LaTroy Hawkins)
- 41: Matt Carpenter (after a -0.3 war season, Ichiro)
- 40: SP Carlos Carrasco (after a 3.9 war season, Bartolo Colon), RP Adam Liberatore (Joe Nathan), 1B Buster Posey (0.6 war)
- 39: 1B Brandon Belt (3.1 war), CL Zach Britton (still closing), RP David Carpenter, RP Aroldis Chapman, RP Zac Rosscup
- 38: RP Madison Bumgarner, RP Patrick Corbin, C Yasmani Grandal, CF Starling Marte, RP Michael Pineda, SP Jose Quintana, 1B Anthony Rizzo, RP Chris Sale

38 or above - game: 19; real life 2015: 25
40 or above - game: 6; real life 2015: 7

It adjusts pretty well. Perhaps we could try a slower aging speed (.290). I tried once .250 and didn't work.

Youngest players in ML (all unknown, so I won't put names)

- 20: 1 ; real life 2015: 3 (Correa, Osuna and Castro)
- 21: 6 ; real life 2015: 14
- 22: 20; real life 2015: 56

So perhaps we have to increase the development speed, from 1.300 to 1.400 and see again

Active players with +500HR: Giancarlo Stanton (588 with 37yrs), +400HR: Joey Gallo (485 with 33 yrs), Mike Trout (466; 36 yrs), A.J.Reed (405; 34 yrs).

HR record breaking chance:
- Joey Gallo if he belts 40HR for 7 more seasons from 34 to 40
- Thomas Johns. 285 HR with 29. 43HR per season for 11 seasons until he's 40. But made 54HR/year last 3 seasons.

Just for fun...

+4 WAR players 30 or above

- Pitchers: 2029 - 3; 2028 - 4; 2027 - 4; real life 2015 - 4
- Batters: 2029 - 17; 2028 - 12; 2027 - 13 real life 2015: 9

[rjl518]

First of all...i want to thank RubeBaker and Sprague for the shout-outs.
I had decided on changing certain settings during the course of a season to see what kind of results i would get with my Mets real time dynasty that is going on.
In my dynasty as of now, My Mets have a record of 30-24.
The actual real life record for the Mets after their loss to the Marlins is 31-24!
I had a rain out earlier in the year which is why i am a game short.
I dont think that i can get any closer than that.
Of course, not all the teams are acting like their RL counterparts.
Even though i had set injuries to LOW to start the season, i still had two key players get hurt early in Michael Conforto and Neil Walker.
So the injury bug still bites.
As far as the trade frequency, now that it has been set to NORMAL for June, the AI is making IMO some decent transactions.
As the season progresses into July where the Trading will be set to HIGH and VERY HIGH, it will be very interesting to see what the playoff bound teams do and what the cellar dweller teams too.
So far, my changes have brought me an interesting season.
Looking forward to the rest of the year.

[PSUColonel]

Do people feel it's necessary to lower scouting accuracy to create a fog of war? Don't players tank enough on the default setting? TCR...what do you guys think about 100 vs. say 150 or 170?

[RubeBaker]

Originally Posted by rjl518

First of all...i want to thank RubeBaker and Sprague for the shout-outs.
I had decided on changing certain settings during the course of a season to see what kind of results i would get with my Mets real time dynasty that is going on.
In my dynasty as of now, My Mets have a record of 30-24.
The actual real life record for the Mets after their loss to the Marlins is 31-24!
I had a rain out earlier in the year which is why i am a game short.
I dont think that i can get any closer than that.
Of course, not all the teams are acting like their RL counterparts.
Even though i had set injuries to LOW to start the season, i still had two key players get hurt early in Michael Conforto and Neil Walker.
So the injury bug still bites.
As far as the trade frequency, now that it has been set to NORMAL for June, the AI is making IMO some decent transactions.
As the season progresses into July where the Trading will be set to HIGH and VERY HIGH, it will be very interesting to see what the playoff bound teams do and what the cellar dweller teams too.
So far, my changes have brought me an interesting season.
Looking forward to the rest of the year.

The injury bug bit me, too. I lost Ben Zobrist for 4 weeks, but Arismendy Alcantara filled that hole admirably. I also lost Jason Heyward for a couple of weeks to the DL although I could have let him stay on the roster for a week. Instead, I chose to put him on the DL and gave Jorge Soler some more playing time. BTW, as of now I am caught up with the real-life Chicago Cubs. Their record is now 39-16, mine is 40-17. Pretty darn close.

[marc5477]

Originally Posted by PSUColonel

Do people feel it's necessary to lower scouting accuracy to create a fog of war? Don't players tank enough on the default setting? TCR...what do you guys think about 100 vs. say 150 or 170?

Negative, on normal settings its waaaaaay to easy to find good prospects and unfortunately, the AI is not as smart. It actually ruined my early games because I eventually ended up having a team of superstars on standby in AAA... it was ridiculous.

The most realistic setting is indeed Very Low. This has been tested by many vets of the game for many years and it is the closest approximation to reality. Using very low, i average maybe 1 superstar every 5-10 years and probably 1 star every other year. I get at least 1 or 2 average or slightly above average player every year which is actually very realistic if you think about it.

The only issue right now with prospects is coaching and how unfriendly the coaching system is. As it turns out, the reputation stats seems to have absolutely no meaning so its just crap shoot when it comes to selecting coaches. The old system was better.

[PSUColonel]

what if you use normal scouting with talent change randomness at 200? ...or do you feel both is more realistic?

[PSUColonel]

I find the problem with very low scouting accuracy to be the over abundance of top teir prospects...the scouts hand out 7 and 8 potential ratings like candy....this too is very unrealistic

[PSUColonel]

Originally Posted by marc5477

Negative, on normal settings its waaaaaay to easy to find good prospects and unfortunately, the AI is not as smart. It actually ruined my early games because I eventually ended up having a team of superstars on standby in AAA... it was ridiculous.

The most realistic setting is indeed Very Low. This has been tested by many vets of the game for many years and it is the closest approximation to reality. Using very low, i average maybe 1 superstar every 5-10 years and probably 1 star every other year. I get at least 1 or 2 average or slightly above average player every year which is actually very realistic if you think about it.

The only issue right now with prospects is coaching and how unfriendly the coaching system is. As it turns out, the reputation stats seems to have absolutely no meaning so its just crap shoot when it comes to selecting coaches. The old system was better.

In watching the draft, and reading some articles, many are saying MLB scouts are actually getting it right a lot more than they have in past years. I would say especially first round picks over the past 8-10 years.

[Bears5122]

These threads are a big help to someone who isn't as knowledgeable in the game. Thanks to all who participate.

Originally Posted by marc5477

Negative, on normal settings its waaaaaay to easy to find good prospects and unfortunately, the AI is not as smart. It actually ruined my early games because I eventually ended up having a team of superstars on standby in AAA... it was ridiculous.

The most realistic setting is indeed Very Low. This has been tested by many vets of the game for many years and it is the closest approximation to reality. Using very low, i average maybe 1 superstar every 5-10 years and probably 1 star every other year. I get at least 1 or 2 average or slightly above average player every year which is actually very realistic if you think about it.

The only issue right now with prospects is coaching and how unfriendly the coaching system is. As it turns out, the reputation stats seems to have absolutely no meaning so its just crap shoot when it comes to selecting coaches. The old system was better.

Would setting it to Low and using a 2-8 scale work just as well? I agree with everything you said but I'm a little nervous on going to Very Low. Almost feel like I have no control over the decisions at that point.

Agree on the coaching ratings too. Although I don't mind it as much because it forces me to focus more on results. Finding quality coaches in real life is kind of a weird crapshoot anyway.

[PSUColonel]

the problem I am finding with very low scouting accuracy, is simply that scouts will rate half the players in the draft 4-5 stars, and it's just ridiculous and not realistic.

I am not happy with this portion of the talent distribution at all. I am going to use TCR at 200 with normal scouting in an effort to create a realistic MLB environment. I have no idea if it will work, but the inflated ratings are just a joke IMO.

[BigRed75]

Originally Posted by PSUColonel

I am not happy with this portion of the talent distribution at all. I am going to use TCR at 200 with normal scouting in an effort to create a realistic MLB environment. I have no idea if it will work, but the inflated ratings are just a joke IMO.

Spanish Lefty ran a lot of tests and simulated leagues and found that a TCR of 170 was the sweet spot with respect to WAR distribution across the league.

[marc5477]

Originally Posted by PSUColonel

the problem I am finding with very low scouting accuracy, is simply that scouts will rate half the players in the draft 4-5 stars, and it's just ridiculous and not realistic.

I am not happy with this portion of the talent distribution at all. I am going to use TCR at 200 with normal scouting in an effort to create a realistic MLB environment. I have no idea if it will work, but the inflated ratings are just a joke IMO.

That is actually how it is in real life. All the kids coming out of HS and Col have "potential" to be great and there is very little difference between anyone after the 1st 20 picks that distinguish them from anyone else. Does that mean that the "other" guys should be rated lower? Dont be too sure since many of the greats came from lower rounds and many high ranked prospects end up fizzling out thus, its clear, that the potential scale has a very wide error range and that is exactly how Marcus modeled it in ootp 17.

As for development randomness, its a matter of flavor. If you set it to 200, it will be as if you were saying that the scouts dont matter nor does experience and skill. Everything is random including the performance of veterans. This has been discussed to length in the past as well and for the most part, I think many will agree that development really isnt that random in reality. If you look at 1st round picks, many do make it to the bigs eventually, albeit most are barely average. Also, established (veteran) superstars and stars dont usually flake out very often and are very consistent for many years (minus injuries) so again, thus it really isnt that random at some point in a players career. Certainly there is some randomness (most acute with younger players), so I just stick with the devs default of 100. I wish there was a separate scale based on age (as age increases, randomness should decrease).

[PSUColonel]

Originally Posted by marc5477

That is actually how it is in real life. All the kids coming out of HS and Col have "potential" to be great and there is very little difference between anyone after the 1st 20 picks that distinguish them from anyone else. Does that mean that the "other" guys should be rated lower? Dont be too sure since many of the greats came from lower rounds and many high ranked prospects end up fizzling out thus, its clear, that the potential scale has a very wide error range and that is exactly how Marcus modeled it in ootp 17.

As for development randomness, its a matter of flavor. If you set it to 200, it will be as if you were saying that the scouts dont matter nor does experience and skill. Everything is random including the performance of veterans. This has been discussed to length in the past as well and for the most part, I think many will agree that development really isnt that random in reality. If you look at 1st round picks, many do make it to the bigs eventually, albeit most are barely average. Also, established (veteran) superstars and stars dont usually flake out very often and are very consistent for many years (minus injuries) so again, thus it really isnt that random at some point in a players career. Certainly there is some randomness (most acute with younger players), so I just stick with the devs default of 100. I wish there was a separate scale based on age (as age increases, randomness should decrease).

They all have great potential, but the number of 7's and 8's given OOTP is very off compared to actual scouting grades...it's actually ridiculous. A top MLB draft prospect might receive something like Hit:55, power:65, eye: 60,,,in a year when a player gets a 7 or 8 in one of those categories, they are considered exceptional, and not seen too often. OOTP reflects this on higher accuracy ratings, and does a horrible job on lower accuracy.

[PSUColonel]

This is just the first page: