PowerPlay Magazine

Greetings all,

It has been awhile since I've written but I'm writing today to let you know of a series of articles coming your way in the near future. This article will serve as a segue into coming articles which aim to make you more educated as to how good your players are compared to the rest of PPM.

Today, I'm going to introduce you to a tool I developed for hockey. This tool will allow you to compare any player of yours to data acquired from systematically searching the PPM abyss. You can input an age, C/L and OR and the spreadsheet will feed back the percentile rating of your player. This can help you make educated decisions about how good players are compared to the rest of PPM teams and based on that decide if they are good enough for your team. For those of you not familiar with how percentile works, if a player is in the 50^th percentile he is in the top 50% of all players. If a player is in the 95^th percentile, he is better than 95% of all players.

To download this tool, you simply need to click HERE (and download the Excel file). However, before you rely on it too heavily, there are some caveats you must consider first.

1.       I systematically fetched player data dating back to day 1 player IDs. I searched every 50 players meaning I am only sampling 2% of the player pool. Given the number of players, this serves as a solid base but is not perfect. There are a few age + C/L combos that do not have good statistics.

2.       There were few 15 year old players sampled by the above resolution. As a result, I improved it to search every 10 players for the higher numbered player IDs. A downside of this is that new teams get players with large player ID numbers. Alas, the age and C/L of players for new teams (18-22 or so) will be skewed towards lower OR compared to developed teams.

3.       I removed all players fetched that were on inactive or noname teams.

In short, this is an approximation tool. It uses a CDF (cumulative distribution function) to extrapolate roughly where your player ranks compared to players of the same age and C/L. You will likely find your player comes in better than you'd expect. However, this is likely an artifact of the caveats mentioned above in the lower age range.

I do have plans to run this script and develop similar tools for the other three sports. Given how nascent basketball is, it is a logical choice to run next as people may want to see how well their day 1 guys (now being caught by rookies) compare to the rest of the database to make educated decisions on whom to keep and whom to sell.

In the coming weeks you will hear from coldmountainrebel on a similar topic. And, our newest editor, Obryantj will be translating an article on the best of the best in hockey - a 18 year old fantasy draft of sorts.

That's all for now. Best wishes from 35,000 feet,

-Scott