In the past few years I have been expressing an increasing affinity for World Football. I gravitated towards AC Milan and Barcelona, as I've always enjoyed the beautiful countries of Spain. Now, how I came to make my decision on the teams I wish to love (Milan and Barca) is too long of a story, but suffice it to say, lots and lots of history was read, lots and lots of games were watched, and lots and lots of reports, trades, administrative files were paged through. It would take a fortnight for me to fully detail exactly how I came to my decision.
Now that I have expressed how I feel about World Football, I should mention that there are a few things about it that I do not particularly enjoy. One of these is the rating system used world wide to rank teams.
My first bitch about this rating system (called AQB Sports Rating) is that the system doesn't take several facets of the game into account; in effect determining the strength of a team by whether or not it wins by big margins. The system neglects the relative strength of the team over time, as well as the relative strength of the league over time. For example, a win of a single powerhouse team in a smaller country's league (say Portugal) over a smaller team known to get beat by 3+ --and, true to form, the powerhouse wins by 5-- clearly outweighs the system. Another facet that is gleefully left out is the main goal scorer and the position that said goal scorer plays. If a team has 1 forward that averages 3 goals a game, and the defense can barely hold off 2, the team isn't strong; yet by the AQB rating system, this team is considered extremely strong.
With this in mind, I chose to devise a new system. I have this new system detailed to its fullest as of now, but it has yet to be proven on determining strengths and weaknesses of teams (as well as certain variables I need to tweak). I will post the Ranking Method --which I have yet to name-- and how to determine each variables as soon as I perfect it. For tonight, it remains unproven, but when I ran some preliminary tests (a terrible team's stats, a mediocre team's stats, and a powerhouse team's stats) I came up with very promising numbers, allowing me to determine which team does well, and which doesn't.
The goal this season is to tweak the numbers a bit and to perfect the system whilst also testing it out.
Friday, July 18, 2008
Subscribe to:
Posts (Atom)
