Quote:
Originally Posted by Steve Yzerman fan
It has a lot of components to it but the basics are that every play has a value attached to it (either positive or negative) and certain types of plays are worth more than others.
Essentially it is an attempt to remove luck from the results. Some plays during a game that would be deemed lucky would be a tipped ball at the line of scrimmage that ends up getting picked or turns into a TD.
Fumble recoveries also play a pretty big role as there is definitely an element of luck involved in those.
Of course the model isn't perfect but I want to see how it works as a potential predictive model. Basically what I've learned is that Toronto should win at a higher rate going forward whereas a team like Winnipeg should lose at a higher rate.
It also tells me that Ottawa and Montreal are almost extreme outliers. Like these two teams could be historically bad whereas the rest of the teams are at least reasonably close.
To expand on this post from a little while back...
To me there seems to be 3 elite teams:
BC
Calgary
Edmonton
1 team that is above average but appears to be "running" poorly:
Toronto
2 teams that are average:
Winnipeg
Saskatchewan
1 team that is bad:
Hamilton
And two teams that are extremely awful:
Ottawa
Montreal
Take Saskatchewan as an example...
They are 5-2 and a lot of people probably think they are a pretty good team but when you take a look at their wins and losses, this is how it shakes down...
Wins:
Winnipeg
Toronto
Hamilton
Ottawa
Montreal
So they are beating the teams that they should beat and also beat a couple of teams that are close to them but no elite teams show up on that list.
Losses:
Toronto
BC
What really stands out to me is the lack of elite teams that they've played against this year. I get kinda scared when I realize how many times they have to play BC, Edmonton and Calgary.