Post by Commish_Ron on Mar 24, 2018 13:21:03 GMT -5
As I mentioned a couple of weeks ago I have been simulating seasons in a Sandbox PBL league to get better understanding of the modifiers. I have simulated 20 seasons and am now ready to make a recommendation.
Please reference the data here:
docs.google.com/spreadsheets/d/1AROxaw5ZESrxi_wWduZbPKAKO4qD3dGN-yENB8kS1Is/edit?usp=sharing
If for some reason you cannot access google docs I also have the data in an Excel spreadsheet that I can share. Or in worse case I can send as a CSV
Across all sims I tracked Plate Appearances, hits, doubles, triples, home runs, runs, walks, strike outs, stolen bases, caught stealing, and the percentage of times each occurred based on plate appearances. My greatest area of interest was home run percentage. In PBL 2043 and 2044 we were at about 2.7%. In 2045 that skyrocketed to over 3.4%. Based on the overwhelming results in a league poll to find a balance between those two extremes my goal was to find settings that would produce 3-3.1% home runs per plate appearances. Obviously also keeping an eye on the other statistics to be sure nothing else was thrown out of balance.
Ok. On to the data and the stories it tells.
At the top for reference are the MLB 2016 actuals that OOTP pulls in to the modifiers settings when you select 2016 as your base year. Then you have the results of the 2043, 2044 and 2045 PBL seasons.
First I ran the 2046 PBL season four times maintaining the current settings. Home run numbers do fall off from the big 2045 boom, but they still stay above the target ranging from 3.20-3.24%
On the settings screen there is a checkbox "automatically adjust league totals modifiers for accuracy." My next run was to sim 2046 using current settings but unchecking that box. HR% exploded to 3.58%. Fail. Moving on.
Theory. If all modifiers are set to 0 and auto adjust is off, shouldn't that create numbers true to actual player ratings? Next run tested that. Epic Fail. League hit percentage dropped to 5.48. On this one I had to take a second to check the league leaders screen because I couldn't believe the numbers. The Batting leader that sim had a .110 average. ROFL. Again, epic fail. Moving on.
Next idea. If MLB base HR of 5610 produces a 3.39 HR%, then mathematically 5132 should produce the 3.0 that I am looking for. Next run was to set base HR to 5132 and uncheck the check box. Another epic fail as this setting produced the highest HR% of any simulation throughout the process. A whopping 3.9%. Moving on.
Next idea. Maybe if all modifiers are set to 1 and auto adjust is unchecked it will produce numbers true to ratings. (Same theory as 2 runs back but with 1 instead of 0). This set was interesting. I ran it once and got another very high HR%. 3.75%. That seemed really bizarre so I ran it two more times. In each of the next two HR% was just above 2.4%. Not sure what happened here. I likely screwed something up on the first run. Regardless, this is a fail as nothing was close to my 3% target.
Next idea. Set base HR to 5132 and hard set modifiers to 1. Only produced 2.6% HR. Fail. Moving on.
Next idea. Create a stock MLB 2016 OOTP league. Note the default out of the box modifier settings. Apply to PBL Sandbox and sim 2046 season. This was a fail. HR% dropped to 2.3%. Also this run only produced 787 Stolen Bases. Out of curiosity I ran a second simulation but adjusted the stolen bases only in an attempt to understand the math and try to correlate setting changes to results. I was trying to force settings to produce 3x as many stolen bases. Instead it produced 13x as many. This cemented that any strategy of attempting to understand how the modifiers work and applying a mathematical equation to implement them is folly.
Next idea. Try out 2015 season base modifiers. This one failed. It still produced 3.6 HR%. But I wanted to try it a second time. At this point I finally got my breakthrough. When I loaded the settings this time I clicked the Auto Adjust check box a couple of times and observed that each time I clicked it, the modifiers adjusted down. I can spam that check box and watch the modifiers drop closer and closer to zero. So I blew away those changes, loaded 2015 default again, clicked auto adjust two times only (to drop the modifiers once. I ran that sim and observed the HR% drop from 3.6 to 2.6%. Missed my mark but finally something that makes some sense!
Finally, that brings me to my recommended settings. What I have done is loaded the 2016 MLB defaults and clicked the auto adjust modifiers off and back on just one time. That generated the settings in the attached screen shots. I have simulated 3 seasons into the future 2 times. In those 6 seasons HR% has ranged 2.8-3.2%. Hits go down some which is a good balance to the increase in HR. Walks are down and strikeouts are up even more. I think that is a good correction. Stolen Base percentage held steady at 71-72%.
This strategy makes sense to me. When clicking that auto adjust it is making across the boards adjustments that will apply changes in a balanced way that any equation I am capable of creating cannot.
I do NOT expect to maintain this 3.0 HR% target long term. What I have done here is come up with recommended settings that I will set for this season, one time. Going forward I would expect and hope the league to evolve organically as talent turns over, new versions of OOTP make adjustments etc. That is all positive evolution in my opinion. I am repeating myself here but I cannot stress it enough. I do not want to be in the business of trying to control league statistics. My entire goal is to be comfortable with a starting point for now. Set once, be done and let what changes come be.
Thoughts?
Please reference the data here:
docs.google.com/spreadsheets/d/1AROxaw5ZESrxi_wWduZbPKAKO4qD3dGN-yENB8kS1Is/edit?usp=sharing
If for some reason you cannot access google docs I also have the data in an Excel spreadsheet that I can share. Or in worse case I can send as a CSV
Across all sims I tracked Plate Appearances, hits, doubles, triples, home runs, runs, walks, strike outs, stolen bases, caught stealing, and the percentage of times each occurred based on plate appearances. My greatest area of interest was home run percentage. In PBL 2043 and 2044 we were at about 2.7%. In 2045 that skyrocketed to over 3.4%. Based on the overwhelming results in a league poll to find a balance between those two extremes my goal was to find settings that would produce 3-3.1% home runs per plate appearances. Obviously also keeping an eye on the other statistics to be sure nothing else was thrown out of balance.
Ok. On to the data and the stories it tells.
At the top for reference are the MLB 2016 actuals that OOTP pulls in to the modifiers settings when you select 2016 as your base year. Then you have the results of the 2043, 2044 and 2045 PBL seasons.
First I ran the 2046 PBL season four times maintaining the current settings. Home run numbers do fall off from the big 2045 boom, but they still stay above the target ranging from 3.20-3.24%
On the settings screen there is a checkbox "automatically adjust league totals modifiers for accuracy." My next run was to sim 2046 using current settings but unchecking that box. HR% exploded to 3.58%. Fail. Moving on.
Theory. If all modifiers are set to 0 and auto adjust is off, shouldn't that create numbers true to actual player ratings? Next run tested that. Epic Fail. League hit percentage dropped to 5.48. On this one I had to take a second to check the league leaders screen because I couldn't believe the numbers. The Batting leader that sim had a .110 average. ROFL. Again, epic fail. Moving on.
Next idea. If MLB base HR of 5610 produces a 3.39 HR%, then mathematically 5132 should produce the 3.0 that I am looking for. Next run was to set base HR to 5132 and uncheck the check box. Another epic fail as this setting produced the highest HR% of any simulation throughout the process. A whopping 3.9%. Moving on.
Next idea. Maybe if all modifiers are set to 1 and auto adjust is unchecked it will produce numbers true to ratings. (Same theory as 2 runs back but with 1 instead of 0). This set was interesting. I ran it once and got another very high HR%. 3.75%. That seemed really bizarre so I ran it two more times. In each of the next two HR% was just above 2.4%. Not sure what happened here. I likely screwed something up on the first run. Regardless, this is a fail as nothing was close to my 3% target.
Next idea. Set base HR to 5132 and hard set modifiers to 1. Only produced 2.6% HR. Fail. Moving on.
Next idea. Create a stock MLB 2016 OOTP league. Note the default out of the box modifier settings. Apply to PBL Sandbox and sim 2046 season. This was a fail. HR% dropped to 2.3%. Also this run only produced 787 Stolen Bases. Out of curiosity I ran a second simulation but adjusted the stolen bases only in an attempt to understand the math and try to correlate setting changes to results. I was trying to force settings to produce 3x as many stolen bases. Instead it produced 13x as many. This cemented that any strategy of attempting to understand how the modifiers work and applying a mathematical equation to implement them is folly.
Next idea. Try out 2015 season base modifiers. This one failed. It still produced 3.6 HR%. But I wanted to try it a second time. At this point I finally got my breakthrough. When I loaded the settings this time I clicked the Auto Adjust check box a couple of times and observed that each time I clicked it, the modifiers adjusted down. I can spam that check box and watch the modifiers drop closer and closer to zero. So I blew away those changes, loaded 2015 default again, clicked auto adjust two times only (to drop the modifiers once. I ran that sim and observed the HR% drop from 3.6 to 2.6%. Missed my mark but finally something that makes some sense!
Finally, that brings me to my recommended settings. What I have done is loaded the 2016 MLB defaults and clicked the auto adjust modifiers off and back on just one time. That generated the settings in the attached screen shots. I have simulated 3 seasons into the future 2 times. In those 6 seasons HR% has ranged 2.8-3.2%. Hits go down some which is a good balance to the increase in HR. Walks are down and strikeouts are up even more. I think that is a good correction. Stolen Base percentage held steady at 71-72%.
This strategy makes sense to me. When clicking that auto adjust it is making across the boards adjustments that will apply changes in a balanced way that any equation I am capable of creating cannot.
I do NOT expect to maintain this 3.0 HR% target long term. What I have done here is come up with recommended settings that I will set for this season, one time. Going forward I would expect and hope the league to evolve organically as talent turns over, new versions of OOTP make adjustments etc. That is all positive evolution in my opinion. I am repeating myself here but I cannot stress it enough. I do not want to be in the business of trying to control league statistics. My entire goal is to be comfortable with a starting point for now. Set once, be done and let what changes come be.
Thoughts?