League Modifers & Stat Outputs
 

Hey guys,

Sorry for starting yet another thread on this. I really need help and its driving me NUTS.

Quick run down - started my first ever fictional league 20 teams (will hit 30 after multiple expansions). Starting the league out in 1980 but looking to start with present day MLB stat totals. I cannot for the life of me get this right.

I selected "2018" for the league totals and it produces way inflated stats. I am not familiar with tweaking this much and for my own sake of trying to self fix it I change one thing and it makes others worse/better.

The defaults the game starts with when making a fictional league are also high compared to present day MLB.

Basically what I am looking for it a set of numbers to input that will produce very similar numbers to present day MLB (few HR hitters of 40 +, high 200's for strike outs... maybe 1-2 peaking right over 300, ERA's between 2 and 3.5, batting average with a handful of people north of .300) basically similar a similar stat(s) output to this:

2018 MLB Statistics

I am very green with any modifier type settings and just cannot get the game to produce a similar stat set. Please please please can someone help :(

EDIT: I've also hit recalc and it just makes things worse/more inflated...

I’m not sure, but is the engine forcing a 20-team league to produce totals that were originally generated by a 30-team league (MLB 2018)?

Thanks for the response pstrickert, and good question... I did some extensive research prior to posting this and there seems to be mixed opinions on that. I am super green with changing any of the modifiers so I have no clue :o I have my fictional league all set to go but I have been messing with this for a few days now and just cannot figure it out.

Its the last thing I need and I will be ready to rock. Hopefully this topic can generate some good responses cause I am lost.

i do not believe the engine works that way. you have enter plate appearances so it scales.

imo, getting a specific league total is incredibly frustrating thing to get down.

During game creation wizard
-new custom game
-create fictional league
-setup 20 team league (i did 2 leagues, 10 teams in each)
-select 2018 or 2019 settings (and untick auto adjust league strategy when advancing)

Once game has loaded
-Go to League Settings > Stats & AI > untick 'Auto adjust league totals modifiers for accuracy'
-advance to first day of spring training
-Go to League Settings > Stats & AI > select '2018' from the 'Totals from year' drop down
-the game will take a few seconds calculating the modifiers
-tick 'Lock League total stats'

Then continue as normal. This works for me in giving pretty accurate 2018 stats output. (I find it is always a little off with HRs with two many HRs going to higher rated players and not enough spread to lower level players but league totals are usually spot on).

I have done that already, and just did it again to test. A lot of the stuff looks pretty good. However it seems the offense Is still a little inflated. Biggest glaring issue in this is HR's. There is a batter with 60+, 5 batters with 50+, and 12 with 40+. Few other things stand out such as hits seem to be a bit high, RBI's are somewhat inflated, etc...

Bump - Anyone else have any info or advice on this topic?

League totals are really complicated because they interact with the talent level in the league to create results. So even when you get them "right" your output will change as the ratings in the league drift. At the end of the day, I'd be surprised if there's an air-tight way to get exactly what you want. Mostly, you kind of tweak things until they're close enough for your patience. :)

But, if you've getting too many homers, just adjust the total up or the modifier down a little and you'll probably get it about right overall.

Note that individual home runs may still not be distributed to your desire. That's going to depend on ratings and rating distribution to some degree.

I wish there were an easier answer. :)

lots here so -- are you choosing to focus on league results or individual results? i have various ways to do both below that are guaranteed to work well. they are not always the same as you can see with the home runs.

autocalculating may fix things too, if you did not autocalculate modifiers... test that and see if it reduces HR... more likely it's because HR total is ~6k? you will see 60-70+ hr guys in that environment fairly often, i would guesstimate. if you don't want that, it is a bit more complicated to fix, but certainly doable.

-------------------

if you want fewer homeruns, you'll have to adjust some other things to make BABIP add up properly after you reduce # of Home runs in total. and any other interrelated state i can't think of off the top of my head.

normally i will suggest always stickign to adjusting modifiers... but, if you plan to make use of autocalculate, you will always have to eyball the adjustment for HRs modifier, since you want fewer than the total allows individuals to reach. autocalc will always recalibrate it to the League Total value.. which are results you don't like, from what you say.

the totals will scale... so, it is of no concern if you use 100K AB and totals relating to 100kAB or 165k AB and totals in proportion to that number of AB etc etc.

so, if you want the same rates of various things league-wide, that is what you will get. however, player distribution in a 20team league and a 30 team league may not be exactly the same curve -- or at least more volatility in the 20T league for sure. meaning you'll have a deeper valley and potentially higher peak years. this may result in elite players getting a larger or smaller chunk when compared to a 30T league etc etc... difficult to know by eyballing it without exact knowledge of all the dynamics invovled in the scaling and creation of players per year.

as always, you'll have a choice to mimic totals or individual results.. the two are not always the same LT/LTM combinations.

i can tell you ~5000 home runs should give you an occasional 60hr guy, but it should not happen very often at all... maybe with some great luck you can see more than i have. i think 2-3 hit 60+ in 150+ years? that's a ~.258ba and ~5000hr league. i use custom totals in every league i create, but they have to add up. they are very similar to defaults, but lower k's lower hr's and a higher BA.

use a spreadshett ot calculate slash and babip from the other stats as you type them in... adjust the other stats to get slash and babip where you want it.

sac flies are not known ahead of time... but, if you stick to to a 30team ~165K ab total, then you can guesstimate this to be 1050-1250 or so (i'd suggest ~1150-1200). you could see this in hindsight after 20-30+ years of collecting data, lol, but not at inception, unfortunately, so some calculatons may be slightly off, but ~50-100 out of 165,500 isn't so bad.

i recommend sticking to the 165500 ab total, or whatever they use, so that you can use RL data to figure out ratio of various things like doubles and triples etc more intuitively. if you choose to adjust these things to focus on individual results rather than league total results, it will make that whole process more familiar for you. one less step to convert to new AB total -- it scales the same no matter what if same ratio/proportions.

no worries of using 30T LTs in a 20T league at all. Do make sure to autocalculate the modifiers once you make any changes to totals, though. if you don't adjust totals, remember to adjust any Modifier after autocalc that you want to deviate... in this case reduce HR by some X%... once you hone that down it should stay consistent reduction any time you auto-calculate the modifiers.

either way, modifiers may take some tweaking over time.. also, fictionally created players are slightly different than what will be created in draft over time... it's better to pre-date league and run it for ~20+ years, then delete historical data and make it seem like it's a brand new league on your start date -- but with a robust FA and nearly all players are from a draft class, not initial seed players -- this also benefits contracts and the entire financial system in general. avoid all the initial filling out of organizations based on AI GMs, so to speak.

if you don't pre-date it is not a cataclysmic effect. Simply autocalculate modifers every few years or so... more often in middle of the transition than at extremes. first 3-5 years should be fine... then maybe every 3 years for a couple times, then back to 3-5 years toward end of first 20-25years.

if you do choose to pre-date, you can start at any date of the calendar year you want... simply delete history and news and it's (nearly) just like a brand new league... in all ways that will matter.

you don't have to start from scratch. hopefully you made a template during creation? start doing that if not right before you "create league". making a template from an existing league is possible too.. i assume it'd take care of B-days with a predate, but if not, you could delete all those players and re-fill the league and MiLs and let it run for 20+ years ("Functions" under League settings - available actions)... all those players will be gone anyway and no different than replacing seed players.

NoOne thank you very much for taking the time to respond and give me a nice detailed post. I am not going to lie this post scared the heck out of me :o This seems way above my knowledge on what it is I "should be doing". I do have a template I created though with all the settings, teams, logos, jerseys, etc that I want. I made the template right after I created the league so it will load all the ballparks and info I want as well. So It will make for easy testing, but beyond that I am clueless.

I am very new to the whole modifiers and what not so I don't know what it is I should really be doing to achieve the stats I am looking for. I can tell you I am looking more so for individual results, I would like to see these totals, averages, etc comparable to that of present day MLB players.

okay simpler -- but you have to do hands-on changes anytime you autocalculate modifiers...

simply reduce resulting HR modifiers by x% each time AFTER and autocalculate. a reduction of ~10% should be roughly a 10% reduction in HR.. it may take soem trial and error and sometimes it may shift a few thousandths this way and that for the best modifier after the deduction too. (not below about volatility... don't trust one year's resulting totals as "bad" or "good", but rather insufficient data to draw any strong conclusions. if it's "too far" off, that'd be a hint, but yo have to have an idea of volatility to make that assumption.

don't autocalculate too often. it will cause very flat results year-to-year.

the other simple thing to understand is that league average and individual results may not both align to what you want to see...

e.g. you want RL 2019ish averages for various stats league wide, but that may result in 70hr players etc... if you care more about individual results, that's how i'd make my decisions on modifiers and any adjustments you want to make with them.

same with preference to league totals averaging out -- just be know there is volatility in results... a "6000" hr league doesn't always reach that high... some down years may be ~5400 or so... or 6600 in an up year... depends on cumulative draft quality for last ~10-20 years. not easy to predict without a lot of attention.

all about your preferences as far as how you attack it... the previous post explains how to do it 1 time, and it's fixed from that point on just with an autocalulate... it seems complicated, but i tried to step through a process to do it... with game up and that thread up to help, i bet you can do it either way. for now, do the simpler.. once you grasp how it works with experience, i bet it becomes apparent to you.

figure most stats can vary ~10% or even more for something like SB attempts. most of the main stats are +/- 10% give or take in a 30team league -- both leagues added together -- important to what i am saying. a smaller league will likely have greater volatility that i cannot guesstimate for you.

I have a spreadsheet I have been using for years to get the league totals I want. I'll PM it to you this evening.

I will have to mess around with what you suggested and see if I can get some satisfying results :):o

Awesome, I would really appreciate that!!! :D

May I also have a copy? I have tried messing with LTM'S with very little success.

i have one in forums posted... search on google to double check this is latest... i haven't changed it much.. but, i think i posted one that handles 2 subleagues and one that handles only 1 league:

http://www.ootpdevelopments.com/boar...d.php?t=264979

if oyu know the basics of spreadheets can you make minor adjustments to first row in work table and "fill" down to fix it... as of now it should add 2 lines at a time per year. (if 2 subleages, highlight top row and empty row before 'filling' down. -- if only one, you have 200 rows, lol -- just highly 1 row at top and fill down an it will be corrected for 1 league sans al/nl type setup.)

this isn't a proper spreadsheet. shouldn't even have to look at the worksheet at all. the input/output page should be all you need after linking the data sheets.

anyway, take the "suggested new LTM" as a very loose guide... they will be accurate if you only change 1 LTM... if you change many, they influence each other.

e.g. if you reduce 3b but not total hits, those 3b will still be 'hits' of some sort, but not 3b. most likely proportional to current LTM outcomes.

e.g. 15 hits, 4 doubles, two 3b and one hr... if i reduce 3b by 1, then 4/15ths or so will go into doubles and 1/15th to hr and the rest to singles. *less to type with smaller numbers.

so, if you are planning on changing hits, double and triples in 1 shot.. expect it to fail miserably based on suggestions.

a newer version should show upper/lower values over time to help guess early, if impatient or don't want to create a suitable sample size to work with. once you know the general volatility to expect... if you see something outside that % you know it can be safely raised or lowered in a way you want.

like i said, maybe google search.. even if not the most recent the basic functions are there.

adding to this table is easy... simply click inside table -- anywhere -- and drag down until # of rows you wish to add (doing this at bottom makes for easier filling/fixign of table -- but you can see through this proces how to change starting year of table too... find the year you want to start and cop to first row, then fill down)... then highlight last row (and maybe an empty one if 2 sub leagues) and "fill" down to end of table... it comes with 100 or so years i think. easy to add as many as you want.

EDIT:

Heh, if they have added or changed the export in any way... it may not work anymore, lol.... it would be apparent, if so... unfortunately i didn't 'link' to teh column headers.. that would be visibly apparent immediately when you see 8000 or a stat that should be <1 etc.... simply correct colum with proper cell address reference and fill down from top -- fixed. should work.. it's worked for years.

also not that my spreadsheet and his won't work well when in a transition from "seed" players to players created for a draft... or if switching from real mlb players as they run out to fictional over time. even if changing to a feeder system etc... anything that will change the outcomes of player distribution will make the data apples and oranges during that time period. it will not stand up over time and require additional work later.

the suggestion i made about pre-dating is essential... or zoom out and write down those ltm from a future set of data with ~normal/uniform generation of players over time. you won't be able to use them until transition is mostly over... a few % of seed or real players leftover probably won't matter much, for example.

best way is to just let it run overnight and through your work day... more is always better (to a point). the smaller the league the more years you need to find the actualy long-term averages that result from your LTM/LT and other settings. ~100 years does well for 30T, in general.. a few more might make it a bit more precise, but i'm not looking for some insane ~99.99999% confidence interval or anything, lol.

Can I have a link to the spreadsheet? I work with Excel so I can play to my heart's content eventually I will figure it out!

My fictional league is in its 10th season(started it with 1952 stats), but I want fewer home runs and better ERAs, so I'm using 1976 stats as my basis. However even with that I find some totals inflated so I continually edit select stats like home runs, bases on balls, etc. to get league totals near where I want them. I've done this in the past with other fictional leagues and it seems to work very well. You just have to keep checking and sometimes do minor readjustments.

Forgot about this last night, so I'll have to link it mid-day today.

yes please do! :D