General Betting
Welcome to Live View – Take the tour to learn more
Start Tour
There is currently 1 person viewing this thread.
I have just created my first passion script that is based on the Dixon & Coles script. I am not going to follow the prices religiously, but use them as a basis to make my bets from.
Does anyone else use a model similar to the Dixon & Coles model? Any success or issues with it?
I was thinking about creating another script that produced prices from shots on target, what do you think of this?
Average number of shots on target in the last 5 premiership seasons that produce a goal. E.G. 3.23 shots on target = 1 goal on average.
Then for each team work out an attack and defence parameter from the amount of shots on goal they have and concede. E.G. 5 shots on target and only 2 conceded equals 1.55 attack (5/3.23) versus 0.62 defence (2/3.23).
What do you think to the above idea?
Does anyone else use a model similar to the Dixon & Coles model? Any success or issues with it?
I was thinking about creating another script that produced prices from shots on target, what do you think of this?
Average number of shots on target in the last 5 premiership seasons that produce a goal. E.G. 3.23 shots on target = 1 goal on average.
Then for each team work out an attack and defence parameter from the amount of shots on goal they have and concede. E.G. 5 shots on target and only 2 conceded equals 1.55 attack (5/3.23) versus 0.62 defence (2/3.23).
What do you think to the above idea?
Show More
I think shots on target in itself is too broad a field. You'd get the more value for someone blasting one from 30 yards into a defender's legs than for an attacker missing an open goal from 6 yards out.
I'm trying to build up a model where I'm designating fields within the scoring area (eg front of goal, middle of area, edge of box etc) and then looking at whether it's a goal, on target, off target, blocked etc. I think getting attacking and defensive stats on both of these could lead to a half decent measure of a team's strength but as with everything the collation of the data is the back breaker.
I'm trying to build up a model where I'm designating fields within the scoring area (eg front of goal, middle of area, edge of box etc) and then looking at whether it's a goal, on target, off target, blocked etc. I think getting attacking and defensive stats on both of these could lead to a half decent measure of a team's strength but as with everything the collation of the data is the back breaker.
Post Your Reply
Please login to post a reply.
Wonder
Instance ID: 13539