|
||||
|
|
OOTP 20 - General Discussions Everything about the newest version of Out of the Park Baseball - officially licensed by MLB.com and the MLBPA. |
|
Thread Tools |
04-02-2019, 04:23 PM | #21 |
Hall Of Famer
Join Date: Dec 2001
Location: Phoenix, AZ
Posts: 3,135
|
So this thread reminded me to go back and take a look at how my league was doing in terms of historical accuracy. Everything seems to be in check EXCEPT SB/1B and SB%...somehow these went from >-10% to now >+10%...SB% is actually >+35%. Questions:
This is a fictional league that current date is 2036. What is it comparing to??? And secondly, I can understand that league talent would impact this number over time but looking back it seemed to have flipped from being way under to now way over the course of 2 seasons. So if I wanted to tweak SB Attempts and SB success modifiers to get back in line with original numbers/values, how much do I bump those numbers (attempts and success) down to get close to %0 variance???
__________________
GM - New Jersey Bears of the NPBL; |
04-02-2019, 09:22 PM | #22 | |
Hall Of Famer
Join Date: Apr 2015
Posts: 7,167
|
Quote:
just a note... sometimes increasing offense a bit can actually lower the ERA of your better pitchers, even though the league ERA rises.. it can, in some circumstances, create a greater deviation from the top talent compared to lesser talent levels. i have an idea of baselines i want, but if bumping it a hair predocues these results, i stick with it (only personal preference, obviously -- i think it's a sweet spot that just opens things up statistically at various ~baseline ranges, not just 1 spot) the totals on their own only create a baseline of sorts... then the modifiers have to make it jive with player talent distribution in your league (maybe not exactly in code and behaviour, but from an end-user standpoint, that's its function) if it's not a transition period, like i mentioned, it will stay within an expected range. i have 150-200+ years of runs that show this type of result. i use static LTM in every league i have created since i bought it for the first time. you can expect +/- 10% or more in any stat. sb attempts are on eof the "more" categories. it takes a long time to get a beat on that long-term average that results from a static set of LTM. if you use autocalculate, instead of one of the 2 spreadsheets spoken of in this thread, you will likely constantly adjust in any ~50 year run or so.. eventually you'd hone it down to the same ~LTM eventually if not reacting too much to volatility... with min.max you can makes some guesses if oyu know a bit about volatility for that stat and a related LTM set (full set, any 1 different can have big effect on another.. fewer hits means fewer sb attempts, all other things remaining the same). Also, if you changed # of HR in LT, did you also adjust BABIP? if not, it's not wise to change the totals and not make them jive mathematically with each other. do the method where i suggest autocalcing, then reduce by some % you've deduced over time to work well. it should remain fairly similar if settings and statistical environment settings are not altered in the meantime. year to year fluctuation, even a decade or 2 of futility can occur, due to dynamics of players created for a draft. it's not always the LTM/LT's cause. koohead -- it's maybe not entire mlb history, but definitely from it. they couldn't compare it to 1871-2018 stats lumped together, i hope... i'd guess broken up by the same eras they have in game -- when LTM change etc... that's a new +/- baseline set of stats? just a guess. you could do the math and compare to that year... maybe it's specific to the year and not average of the era? it will, nonetheless be something similar. that screen is only visible when you start with a real world league, i think. even if you transition to fictional after 2019 and beyond. so, not sure waht the baseline they choose..maybe 2018 in perpetuity? again, if you do the math and compare you'd see if it's 35% above some RL result or not. sb attempts... just volatile as heck even with static LTM. don't knee-jerk react to any one year... if oyu think it drops too low or high, make and adjustment due to peak or valley of concern. i'd focus on individuals for this particular LTM -- are the best guys getting the #s you want/expect? don't worry about the league-wide stuff on that. if oyu want a guy to reach 100 with max speed/stealing then up it until that occurs at the rate you want to see. it's funky changing sb-attempts and success... once can alter the range of the other etc etc... you'll see it has multiple "steps" or if diagrmed "U" or rightside of U-shape shapes and hypotenuses inbetween (spelling) or, that's my guess. you can rais it incrementally while changin nothign else and it will go up, up, and then down significantly (100+years of data averaged, not 1 year) more than a few ltm work this way, but not all. Last edited by NoOne; 04-02-2019 at 09:34 PM. |
|
04-02-2019, 09:26 PM | #23 | |
Hall Of Famer
Join Date: Apr 2015
Posts: 7,167
|
Quote:
if you refer to my post, it's in that thread i linked. at very bottom it has a zip-file for an ODS spreadsheet -- i re-attached, but check out the screenies in other thread. there are screenshots of the input/output page and quasi-worksheet. i didn't bother screenshotting the linked data sheets. you simply have to have the exported csv files in same folder as the spreadsheet, then update the links. all data is refreshed immediately. really they don't have to be same directory, you could link to them in the export directory of your saved game too... simply point them there. (google "link data files worksheet" or "link file worksheet" for instuctions for various spreadsheet programs -- excel, libre, openoffice will be very similar but maybe slightly different process) it was made with openoffice calc, but loads up fine in libreoffice and excell to... or it should. oh, and you can jsut "hide all comments" to get that crap out of the way... or delete them etc. this is from 2016, i'll see if it still works when i make a league in XX. lol... really simple to fix on your own too. in my open office, i make record a macro for updating these links... i simply click a button in my toolbar and it updates allf iles. typically, a spreadsheet program will defualt to ask you to update any linked tables... but if you changed settings or maybe excel doesn't do this etc... most likely any time you open, it will ask to update... just make sure exported files exist and are where they should be. click yes on open, and all is right in the world. Last edited by NoOne; 04-02-2019 at 09:42 PM. |
|
04-02-2019, 10:22 PM | #24 |
Hall Of Famer
Join Date: Dec 2001
Location: SF Area, California Total Posts: 531,691
Posts: 2,363
|
Here is the file I use.
Basically, enter your desired stats into column B of the Test tab. Enter your Modifiers in column D. Run a sim (backup your league!) and enter the stats output into column E. It will auto-calc what the next Modifier should be set at (in column H). Proceed right until you get the stats you want. I have Conditional Formatting set for red (too high) at >2% and yellow (too low) at <2%. You can set these how you want by editing the Conditional Formatting setting. Note: there can be a kind of a pendulum effect for lower number stats, so keep an eye out for that. I also have minor leagues set to 30 teams, so you may want to adjust those if your minor leagues are less than 30 teams so your minors stats output doesn't come out wonky. I use this for a heavily modified league, but I haven't needed it for other OOTP quickstarts. Let me know if you have any questions.
__________________
JML MILKSHAKES Last edited by Rizon; 04-02-2019 at 10:25 PM. |
04-02-2019, 11:20 PM | #25 |
Hall Of Famer
Join Date: May 2012
Location: Michigan
Posts: 2,541
|
Thanks NoOne and Rizon,
Sadly these spreadsheets are far to confusing for me to use I don't even know where to begin on either one. Also putting in league totals is a bit confusing for me personally since I have been trying to achieve player stat totals and not league so it would be a lot of guess work for me. I guess I will just have to settle for out of the box at this point or just tweak around until I come up with something that works.
__________________
|
04-03-2019, 01:11 PM | #26 | |
Hall Of Famer
Join Date: Dec 2001
Location: Phoenix, AZ
Posts: 3,135
|
Quote:
__________________
GM - New Jersey Bears of the NPBL; |
|
04-03-2019, 01:19 PM | #27 | |
All Star Starter
Join Date: Sep 2015
Location: Kelowna, British Columbia
Posts: 1,258
|
Quote:
Thank you very much, I can play to my hearts content, I hope you do not regret asking me if I have any questions . If I become a PITA just let me know..... |
|
04-03-2019, 01:21 PM | #28 | |
All Star Starter
Join Date: Sep 2015
Location: Kelowna, British Columbia
Posts: 1,258
|
Quote:
|
|
04-03-2019, 01:42 PM | #29 |
All Star Reserve
Join Date: Oct 2003
Posts: 733
|
did you by any chance move the bases closer together? that will increase your SB and success rates
|
04-03-2019, 02:56 PM | #30 | |
Hall Of Famer
Join Date: Dec 2001
Location: Phoenix, AZ
Posts: 3,135
|
Quote:
oy vey
__________________
GM - New Jersey Bears of the NPBL; |
|
04-03-2019, 03:19 PM | #31 | |
Hall Of Famer
Join Date: Dec 2001
Location: SF Area, California Total Posts: 531,691
Posts: 2,363
|
Quote:
__________________
JML MILKSHAKES |
|
04-04-2019, 02:29 AM | #32 | |
Hall Of Famer
Join Date: Apr 2015
Posts: 7,167
|
Quote:
i bet ~ +/-15-20% from average (or a range =30-40% average, i.e. HUGE fluctuation possible with stolen base attempts, league wide) from whatever all other things jiving together in your context result in -- with static settings/LTM/LT etc. sb % success will probably have a 4-5% total range or +/- 2-2.5% now that's also assuming some baseline near RL... no idea what the range of deviation would be for a ~50% success rate... it could influence dynamic, or it may not... i haven't compared things ike that, i stick to a similar range of baselines each year. this is also based entirely on experience with a 30t 162g schedule... if less, expect more volatility in 1 year than what i say above. if more TeamsxGames, then expect less valotility in 1 year's results. if you are lookign at AL and NL totals, that's 1/2 the size of what i am speaking about above... so totally different context and much greater volatility. you probably wont get everythign you want if you have very specific wants in this aspect relative to distribtion at each talent level possible. so, try to focus on what you care most about. some middling guys or HoF guys? or whatever you wish... dial back expectations if you can't reach them given the tools. maybe it's what you hate most and want to avoid... all depends. |
|
04-04-2019, 02:33 AM | #33 | |
Hall Of Famer
Join Date: Apr 2015
Posts: 7,167
|
Quote:
there's at least one person that does everythign per 100,000ab regardless of league size. when i opened that spredsheet of mine from attachment, i noticed it was using 100k ab... i tried that one year. i've even tried 1,000,000 thinking itmay give more resolution, but it really doesn't as long as not too small. the game has it's max resolution, no idea how many AB to reach that ceiling and anything above is irrelevant. it's all proportional and scales. no worries on that. maybe, in extreme contexts it may skew the curve? i.e. make a league with a .001 BA lol... some funky things are bound to happen with distribution due to insane volatility, i'd assume and a drastic reductions in differentiation from elite and poor quality ratings related to hitting a ball (and the pitcher side of that coin too). Last edited by NoOne; 04-04-2019 at 02:35 AM. |
|
04-04-2019, 02:50 AM | #34 | |
Hall Of Famer
Join Date: Apr 2015
Posts: 7,167
|
Quote:
check out my sheet, it's going to save you time and carpal tunnel syndrome. jaburns, Hockey13Playa go to the thread i linked above, it has a long drawn out process that may have tidbits to help with the actual spreadsheet useage -- getting the data in there etc... you need to export stats -- customize export data so you only get the batting, pitching and fielding csv.. make sure "headers" are inlcuded -- in the customization start be "deselect all" then check those 4 boxes i mentioned. as i mentioned before, my spreadsheet will only work out-of-the box with aleague that has 2 sub-leagues -- like al and nl in mlb, but name clearly isn't relevant. (# of teams or gp is irrelevant, it'll work in all of those contexts). fixing this is easy. it involves "fixing" the first row of the "worksheet" data table.. then, "filling" down to bottom of table. all fixed in a few clicks. basically you just stop it from adding 2 rows together, then fill that down... will 2x the data's rows too 200 years ready instead of 100... extra is no problem... too few of rows is a problem. now you export the data... they show up in import/export folder. per year subfolder? you'll find it. you can even set this up to automatically occur each year -- on jan 1 it'll dump the data. now put those three files in same directory of the spreadsheet for ease -- doesn't have to be there.. can be literally anywhere... whatever you do, be consistent. if yo move them, always move them -- that way you don't have to update the path to the file itself when you update data for the spreadsheet. if unfamiliar with linking to a table from file: https://www.google.com/search?client...ta+spreadsheet google similar search terms. when you open my spreadsheet, the last three worksheets are batting, pitching and fielding -- you never touch these or you will break it! lol. when you update links to these 3 exported files, they will fill and replace any existing data on those worksheets. no work from you, beside clicking "yes" update, please. no typing in data necessary. in future, you simply update the data links and it is done. all stuff is re-calculated immediately (assuming you didn't change that default setting in the program itself, lol.. think f3 or f5 refreshes if you need to manually do it.) i have the path for the same directoy of the spreadhseet, if not mistaken, so best to do that if you don't want to fiddle with it. just dump them in same directory before you load spreadhseet. when it opens it will ask you to update the data links [sic].. simply say yes and it's done. your historica data will be inputed and all calculations will immediately be refreshed. that's all you ever have to do... dump new csv exports into directory, and open the program, then read the input/output worksheet. if oyu want to excludes some year's data, just go remove it from worksheet table -- again, never touch the batting, fielding or pitching worksheets. the other thread has instrucitons on how to shift the starting point of data table on "worksheet" worksheet, and how to resize it to any # of years you want. (it only fills with data, so 100rows and 10 years is no problem.. .but if oyu have 1000years, the the table must be expanded, and re-filled down. instructions in other thread and google.) EDIT: so, i previously used openoffice calc and now have switched to libreoffice just this week for better conditional formatting options due to other spreadsheets i have made and use... it's a bit different defaults, so i want to show this snapshot: as you can see thre is a mesasge at top that auto-updating is disabled in this spreadsheet -- default that i did not knwo and idfferent than my openoffice calc. i will have to enable this option with the button that is visible, then each time i load in future it will always ask to update. even though slightly different, google will get you through this if what i say above doesn't match excel or libreoffice calc. obviously read-only mode is definitely not conducive to using this spreadsheet, lol, i have to ensure that is fixed too. best to use openoffice calc for least problems, lol... if i fix it i'll re-up... i haven't started on a serious XXleague, yet. but, i will use the spreadsheet to make some LTM at some point. Last edited by NoOne; 04-04-2019 at 03:07 AM. |
|
Bookmarks |
Thread Tools | |
|
|