Back and lay
Just to clarify are looking for the total that has been matched during 60secs, 5mins etc?
-
ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
They are, but it tends to be measured from the perspective of the side taking the bet rather than offering. Eg If there's a steamer people tend to say it's being backed rather than layed.
-
firlandsfarm
- Posts: 2720
- Joined: Sat May 03, 2014 8:20 am
I guess you could capture the market amounts at timed intervals and extract the amount over a time period but that won't tell you if the matches were driven by backers or layers nor how much of the money is 'genuine' backing and laying and how much is traders manipulating positions.
This takes the £ Matched over 5 secs but could be any time period them multiplies that value by the WOM % to split it into 2 distinct streams of £. £/s might be a metric to look at.
The idea being to record this data and look at the values of the odds / £ when they move. Recorded a few already but not looked at any.
The idea being to record this data and look at the values of the odds / £ when they move. Recorded a few already but not looked at any.
You do not have the required permissions to view the files attached to this post.
-
ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
WoM moves too rapidly to be much use in allocating the side. You need to use a method similar to deriving PIQ. I did an excel sheet to allocate money to one side or the other which attached to BA a couple of years ago and posted about the issues.Atho55 wrote: ↑Tue Jun 23, 2020 2:55 pmThis takes the £ Matched over 5 secs but could be any time period them multiplies that value by the WOM % to split it into 2 distinct streams of £. £/s might be a metric to look at.
The idea being to record this data and look at the values of the odds / £ when they move. Recorded a few already but not looked at any.
WOM Split.jpg
It basically used LTP and selection vol change to guess what side the bets were but the data conflation means its always an approximation. I used a live pie chart as a swingometer but you've got all the usual issues about what timescale you look at. Eg all day or just the last 3seconds. I ended up moving it to my own software and the tick data to improve it further but it didn't look very useful beyond a few hundered milliseconds ahead.
Tbh money taken is only half the story, this is just a simple game of supply and demand so supply, ie offers cancelled and added are just as important. Possibly more so because people offering and those taking generally operate at different levels of shrewedness imo.
Why not replicate or post the spreadsheet you have already done and see if some forum brain power can be applied to it.
As for my own spin on it, who`s to say what will or not work as long as any change in later review of the data can be related to odds movement signals. The value of the numbers is meaningless but the underlying mechanism should be a constant to drive any change.
As for my own spin on it, who`s to say what will or not work as long as any change in later review of the data can be related to odds movement signals. The value of the numbers is meaningless but the underlying mechanism should be a constant to drive any change.