Player Analysis

[Hoolix]

Listed below is how I'm analyzing my data. Let me know if you think I should be trying to do it some other way that may be better. It does a pretty good job of aligning with who I think had the best season or who should be the highest-ranked player at a position.

Pre-season Rankings



For RF the formula is the average of OVR + Con*.9 + Gap*.25 + Pow + Eye + AvoidK*.9 + Spd*.25 + Stl*.05 + BR*.1 + Spd*.15 + OfRng * .85 + OfErr + OfArm*.75.

Each position has slightly different modifiers based on what I deem essential for that position.

Post Season Stats



For RF the formula is ((OPS + wRC+)* wOBA + wRAA*.25)*.7 + (RNG*EFF + ZR)*.3.

This was the first year that I lumped defense into ranking my hitters. It's not anywhere near perfect and would love some feedback on what I could do to hone in my rankings.

*The .7 and .3 is so that offense (for the RF) is worth 70% of the score. Other positions (SS/CF) have higher defensive %

**The modifiers in the first formula come almost exclusively from GMCheatSheets.

***The modifiers in the Post Season Rankings are just me trying different formulas and trying to figure out something that encompasses the different aspects of the game.

[uruguru]

I recently did a regression on the five offensive ratings (contact, gap, power, eye, avoid K) to correlate them to the projected OBP and SLG in the player editor. My goal was to be able to use those ratings to create a ranking of players by expected offensive output. I found a model that resulted in standard error of .005 for predicting OBP and .006 for SLG. That's about as close as I could get it so I just gave up on it as I was hoping for something with tighter bars. It was a fun exercise. The formulas look complicated but are trivial to use in a spreadsheet application.