Sorry...
I like the one that says "Output for league matches selected year."
Quote:
|
You've given a reason not to use more than one year. Is there a reason to use it?
|
Have less players with limited stats to calculate ratings from, smooth out performance extremes by incorporating more data, etc...
By the above I don't mean that everyone would want to use 3 or 5 year because of this but for some players these reasons are desirable. If your driving goal is to replicate a season as much as possible, than you probably want to stick with 1 year. If you want to limit players getting rated with limited data or limited the impact of extreme single season performance than go with 3 or 5 year.
I've used single and multi year recalc in my different solo leagues. I don't like playing with as played lineups because I prefer to be involved in as many ways as I can, so I don't like it when there are players based on low AB/IP because I don't want to abuse highly rated players who only played a little bit. If I used as played lineups (which I have), then I don't mind using players with limited ABs because the as played lineups helps keep their usage under control.
When I use multi year recalc I tend to use the double the weight of the current year setting to skew the performance towards the season being played.
I've also played games with no recalc where players develop according to the development engine.
I guess my point is that the different settings all have their situations where they are desirable, but some people will never find a certain setting appealing. Anyone who wants to have their players' performances to most closely match a particular season probably will never want multiyear recalc, unless they have other factors that are more important to them.