[One Great Matrix]

Originally Posted by allenciox

Interesting question, if I understand it correctly. So, first of all, the average home run percentage, ba, etc. is going to be based on the reference year. For most of PT play, that year is 2010. If they changed it to say, 2019 but used exactly the same players, then home runs would increase significantly since there were many more home runs hit per ab last year than in 2010.

But you might be asking something different, and that is the more interesting question in my mind --- how does rating variance play into things? What is the difference between having all players with the same rating for a stat versus a wide range of values? And I think the answer is that it makes a difference in terms of how important that stat and its "oppositional" stat is. Let me explain.

I did data analysis on hundreds of tournaments (collectively) between bronze, silver, gold, etc. and the importance of stats was signficantly different as you moved up the levels.

Take avoidK, for example. In a baseline universe, CON alone should be the best predictor of batting average, and in all environments it does a "decent" job of predicting it. But CON is a derived stat from POW, BABIP, and AvoidK rating, and in some tournament types, it has company.

In bronze tournaments, doing a multiple regression to predict batting average shows that both POW and AvoidK need to be added to get the best prediction --- on average, the higher the POW, the higher the BA, and the lower the AvoidK, the higher the BA, holding CON constant.

In silver tournaments, the best predictor of batting average is CON by itself, adding POW and AvoidK to the mix adds no additional predictability.

But by the time you get to diamond tournaments, POW and AvoidK once again enter the equation, but with the exact opposite effects that they have for bronze tournaments--- holding CON constant, the higher the POW the lower the batting average, and the higher the AvoidK the higher the batting average.

So what is going on? I think (but don't know) that the answer is variance. Let's compare bronze players and diamond players. There are very few diamond players with low power, almost all of them have POW > 50, and most are much higher. Among bronze players there is considerably more variance in POW. Similarly, for pitchers, the STU rating for almost all diamond pitchers is quite high, for bronze pitchers there is more variance. Thoughts?

In 2.0 I did a week in diamond last week & I noticed that an ideal squad was a little different...which gave me some ideas about this week back in perfects.

In a nutshell, I don't think this is something we figure out (or when we do the game's over), I think it's the way baseball is... lot of different possibilities, even with ...well, just a lot of different possibilities. I'm sure...I've noticed that much.

Being in Diamond gave me some ideas about this week's perfect level team but mostly I felt like I'm getting it right on the fly and that's why I even bothered to adjust my team when I was playing in Diamond (they won 124 of their games).

I guess what it did for Perfect was mostly remind me that I was in a league with a dazzling array of talent. So...you might think I tried to figure out who my very best players were... But no. Possibly no such thing. Or...at least not for me...this week. Because...that is basically what I did to win a Perfect League, was keep perfecting my team...did take me awhile, but I did repeat the feat.

Instead I ran Hal Brown out there in the rotation. He's got a great control rating. Slightly below average in stuff and movement... His ERA is in the 5's but he's holding his own not giving out a bunch of free passes & has...a slightly better record than Tom Seaver for the time being. I think he is 6-4. Much credit to the offense as he hasn't pitched any kind of real gem yet but sort of limits the damage with the pinpoint control.