I still like to tweak my formula, but since there are a few surprises (at first glance) in the top ten this week I thought I would post a little more detail/philosophy.
My overriding philosophy in my rankings is to rank the teams from top to bottom in Major League Soccer on who is playing best at that point in time based on a formula. I am not trying to predict how the season will end with these rankings. Theoretically, the top team in my rankings on any given week should be able to beat the other seventeen on a neutral field that given week, the second team should beat the sixteen below them, etc.
I don't purely go by this when I make my MLS match predictions, as I look at each teams' ranking score in my formula and their differential to determine whether the match can be predicted. So if the #1 team's score in my rankings has a very small differential with #2, I wouldn't necessarily pick the #1 team to win. I also take into account the predominant home field advantage that historically exists in the MLS to make my pick.
So what is my formula? I take season to date points
per game (not goals, but points earned) as I feel teams should be rewarded for
their long term performance in my rankings. I combine that with points
per game in a team's last five matches (a reflection of who is currently hot)
to come up with a team's score. I weight season-long performance 60% in my
formula and the last five matches' performance 40%. In the last five matches'
average I give a higher weighting for away wins, and will punish a team for a
home loss.
The LA Galaxy are #1 |
1) LA Galaxy (West) |
2) Real Salt Lake (West) |
3) Seattle Sounders FC (West) |
4) Portland Timbers (West) |
5) Toronto FC (East) |
6) Chicago Fire (East) |
7) D.C. United (East) |
8) Houston Dynamo (East) |
9) New York Red Bulls (East) |
10) Sporting Kansas City (East) |
11) Philadelphia Union (East) |
12) Colorado Rapids (West) |
13) FC Dallas (West) |
14) Columbus Crew (East) |
15) New England Revolution (East) |
16) San Jose Earthquakes (West) |
17) Vancouver Whitecaps (West) |
18) Chivas USA (West) |
At first glance, I'm thinking that readers are going to look at this week's rankings and are going to wonder why two teams with a slim to none chance at making the MLS Playoffs, the Chicago Fire and Toronto FC are ranked so high, and virtual locks for the playoffs, FC Dallas and the Columbus Crew are ranked so low.
The reason is current form (last five matches) and it's 40% weighting in my formula. Here are the teams ranked by current form this week:
1) Real Salt Lake (West) |
2) LA Galaxy (West) |
3) Toronto FC (East) |
4) (3-way tie) Seattle Sounders FC (West) |
Portland Timbers (West) |
Chicago Fire (East) |
7) D.C. United (East) |
8) (2-way tie) Houston Dynamo (East) |
New York Red Bulls (East) |
10) (2-way tie) Sporting Kansas City (East) |
New England Revolution (East) |
12) (2-way tie) Philadelphia Union (East) |
Colorado Rapids (West) |
14) Vancouver Whitecaps (West) |
15) (3-way tie) FC Dallas (West) |
Columbus Crew (East) |
San Jose Earthquakes (West) |
18) Chivas USA (West) |
Toronto FC (last 5, WLTWW) and the Chicago Fire
(TWWLW) are playing very well right now and are ranked in the top six in MLS
for their last five matches. FC Dallas (TLWLL) and the Columbus Crew (WLLTL)
are currently playing very poorly and are ranked in the bottom four.
I have admittedly questioned the 40% weighting on
current form, but if I'm going to use these rankings to predict match results,
I feel I have to look at who's hot and take that into account.
So that's my explanation. Not perfect, but it entertains me.
No comments:
Post a Comment