Fantastic website and the best one for Yahoo fantasy sports I have ever seen. Truly phenomenal ground breaking slay work. A few suggestions for player ratings page:
1) a "mean" z-score for each player across every category (used as a shorthand for that player's value - can be used in rankings, tables, and easy predictions for upcoming weeks)
2) turnover z-score should be inversed (high turnovers = low z score) to reflect that low turnovers are more valuable
3) FG% and FT% are tricky.. and using the distribution of percentages of each player without taking into account volume leads to less useful statistic. They should both be calculated as follows:
1. calculate sum makes (Xmakes) and sum misses (Xmisses) from all players combined
2. let _Xefficiency_ = _Xmakes_ / (_Xmisses_ + _Xmakes_)
3. for player P: let _Pefficiency_ = _Pmakes_ /( _Pmakes_ + _Pmisses_ ) where _Pmakes_ is mean makes per game and _Pmisses_ is mean misses per game
4. for player P: let _Pdiff_ = _Pefficiency_ - _Xefficiency_
5. for player P: let _Padj_ = _Pdiff_ * (_Pmakes_ + _Pmisses_)
6. Calculate Zscore for each player based on distribution of _Padj_s of all players
Fantastic website and the best one for Yahoo fantasy sports I have ever seen. Truly phenomenal ground breaking slay work. A few suggestions for player ratings page:
1) a "mean" z-score for each player across every category (used as a shorthand for that player's value - can be used in rankings, tables, and easy predictions for upcoming weeks)
2) turnover z-score should be inversed (high turnovers = low z score) to reflect that low turnovers are more valuable
3) FG% and FT% are tricky.. and using the distribution of percentages of each player without taking into account volume leads to less useful statistic. They should both be calculated as follows: