September 3, 2014

CFL Simulation Results: Who will end the year on top?

CFL.ca

CFL.ca Staff

TORONTO — A computer simulation model is used to calculate the probabilities of any CFL team finishing first in its division.   

The model considers the following:
•             Each team’s current win-loss record
•             Opponents already played (including whether it was a home or away game)
•             Margin of victory (or loss) in games previously played in the season
•             Remaining opponents to be played (including whether those games are home or away)  

The model calculates each team’s probability of victory in each remaining game.  It then simulates 10,000 replications of the remaining regular season schedule.  The first place team in each division is the one with the most regular season wins.  For each replication, the model keeps track of which team finished first.  For instance, if Winnipeg finished first in the Western Division in 1,990 of the 10,000 replications, then its first place probability is 1,990 / 10,000 = 19.90%.

The model is updated weekly based on the results of games played that week.

Week 12 First Place Probabilities
West Division
Team Probability
Calgary 94.70%
Saskatchewan 2.35%
Edmonton 2.23%
BC .71%%
Winnipeg .01%
East Divison
Team Probability
Hamilton 55.9%
Toronto 36.3%
Montreal 7.4%
Ottawa 0.4%
Week 11 First Place Probabilities
West Division
Team Probability
Calgary 89.32%
Saskatchewan 7.8%
Edmonton 2.24%
BC 0.63%
Winnipeg 0.02%
East Divison
Team Probability
Toronto 51.9%
Hamilton 36.1%
Montreal 11.1%
Ottawa 0.9%
Week 10 First Place Probabilities
West Division
Team Probability
Calgary 80.0%
Edmonton 12.0%
Saskatchewan 7.0%
BC 0.7%
Winnipeg 0.3%
East Divison
Team Probability
Hamilton 51.1%
Toronto 42.2%
Montreal 5.6%
Ottawa 1.1%