What is the best way to reverse this trend?

[actionjackson]

Originally Posted by Tyler87898

So I see in another one of your threads that you aren't a fan of using autocalc every year, but it sounds like it would solve my problems, right?

edit: Maybe I should do autocalc each year, but first set all the modifiers to 1.000 before I hit the button.

When you do auto-calc, you allow the game to set the modifiers so that it can achieve the league totals you are hoping to achieve. The reason NoOne doesn't care for it, is that it sims your league three times, and then spits out the LTMs that will be needed to achieve the league totals you're looking for. He feels that's not enough (he's probably right) to achieve his level of statistical accuracy, but for me it comes close enough. If I look at the totals that occur over time, it averages out to roughly where it should.

For example, I use 1984 as the base for my league totals. In 1984, the batting average in MLB was .260. Through 30 years in my league, it is .260. It has fluctuated depending on talent levels in my two sub-leagues, but that's pretty darn good IMHO. In 1984, the earned run average in MLB was 3.81. Through 30 years in my league, it is 3.83. Again, there have been fluctuations depending on talent in the two sub-leagues, but that's a pretty good result. Good enough for me to stick with it. Here's a picture of the fluctuations in action so you get an idea of what I'm talking about:

[actionjackson]

Originally Posted by actionjackson

When you do auto-calc, you allow the game to set the modifiers so that it can achieve the league totals you are hoping to achieve. The reason NoOne doesn't care for it, is that it sims your league three times, and then spits out the LTMs that will be needed to achieve the league totals you're looking for. He feels that's not enough (he's probably right) to achieve his level of statistical accuracy, but for me it comes close enough. If I look at the totals that occur over time, it averages out to roughly where it should.

For example, I use 1984 as the base for my league totals. In 1984, the batting average in MLB was .260. Through 30 years in my league, it is .260. It has fluctuated depending on talent levels in my two sub-leagues, but that's pretty darn good IMHO. In 1984, the earned run average in MLB was 3.81. Through 30 years in my league, it is 3.83. Again, there have been fluctuations depending on talent in the two sub-leagues, but that's a pretty good result. Good enough for me to stick with it. Here's a picture of the fluctuations in action so you get an idea of what I'm talking about:

As you can see, the highest sub-league batting average during this period is .270 in the AL in 1926, while the lowest sub-league batting average is .252 in the NL in 1907. The highest sub-league ERA during this period is 4.24 in the AL in 1902, while the lowest sub-league ERA is 3.47 in the NL in 1907. A decent amount of variance, but it does average out to roughly what I want.

[Tyler87898]

Well, I'm all for a bit of variance, as long as it doesn't get out of hand. I think autocalc is the answer to my problem, and I'm glad I understand it a bit better now.

[actionjackson]

Originally Posted by Tyler87898

Well, I'm all for a bit of variance, as long as it doesn't get out of hand. I think autocalc is the answer to my problem, and I'm glad I understand it a bit better now.

The key is to do auto-calc once a year on the day before Opening Day, so that teams in the game have basically made all of their free agent signings and trades and have their teams together, as much as the game can account for them. Then the game sims the season three times and spits out the LTMs to attain the numbers for whatever season you're trying to copy. I use 1984 myself, but you can use whatever season you want. I'd recommend steering clear of the steroid years and the 1920's and 1930's when they did something to the ball, as it sounds like you've had enough of the high offensive numbers that you're seeing.

Also all I do when I use auto-calc is input the year 1984 in the space for the season. After that auto-calc does the rest. I don't touch the Auto-Calc button itself because I don't like the numbers I've seen when I do. I also have my PCMs and Strategy Settings set to 1984, but that's just a personal preference as I like to have PCMs, LTMs and Strategy Settings all line up as they would if I were playing a straight historical. Remember YMMV, and keep experimenting with it. Glad to see you feel a bit better about things then when this thread and the other one started.

[NoOne]

in tyler's case, alot of this won't apply, because he is having abnormal statistical results (loosely speaking to avoid potential arguments about technicalities and symantics). in his case, autocalc will be making a significant change to make an obvious correction in output.

that's half of it - sample size. 3 yeas is okay, if the current talent levels in the league are close to average. even then, randomness will play a noticeable role in results. also, as i pointed out in a previous post in this thread, if talent is high in your league and LTMs are lowered to reach the inputed value for the League Total, you're actually reducing statistics of players that are rated to perform better. they are the victims of a coincidence that can be avoided.

the other half is that it alters likelyhood of this or that happening instead of it being an even playing field each year. e.g. if you increase home run LTM, those playes in those years have a greater probability of hitting a homerun relative to their power rating. that same power rating in other years equates to fewer home runs. imo, fluctuation should be caused by rule changes, player talent, and other tangible factors the game accounts for.

with a baseline environment, i can say for a fact that player A in 2015 was a better HR hitter than player B in 2088 (again apply common sense to save typing - any metric of your choosing as to why they are better/worse).

wherease in an environment where LTM change constantly and player A was fortunate enough to play most of his years when the league was low in power talent and benefited from the elevated LTM for HR, while player B's career mostly had the opposite effect and likely lowered his HR total, all other things remaining the same.

that type of stuff eats me up. in the real world there is causality. the physics of gravity and friction coefficient of the air are the same as they were when baseball started. what has changed? players workout, peds, better strategies based on better data collection and mining etc etc. tangible things cause change in the real world. simulations are an attempt to model the real world or some aspect of it. therefore, tangible things should cause change in the simulation.

don't get that confused with never changing an LTM. that is not what was stated in any way. there merely has to be a reason to do it. e.g. the game doesn't know what heighth the mound is... if you raise or lower the mound n the game, the only way to logically represent that in the simulation is to change the LTMs. could you change PCMs and get the same result? sure, but is that what happens in real life? do they raise the mound and that action causes players to get weaker and their eye gets worse? nope. causality is the key. in this cause it's about physics and leverage blah blah blah - adding that to the game isn't worth it, lol.