I started a thread on another forum once, called "DNA of a Longshot Winner". It was hugely popular with 56k views and 870 posts. It was shut down in the end, allegedly it was so big, peoples computers slowed on opening it!...lol
This is post is about the DNA of the last three Betfair matched prices. DNA in science is a molecule that carries genetic instructions. I have represented that with a ribbon in the attachment. Each cell is unique and identifies one of the 27 possible 'instructions' given by the last three matched prices.
Comprises: "0"..current matched price, same as last.
"1"..current matched price > last & "-1"..current matched price < last. These are concatenated to form a trinary logic value. Now I'm not going to say anymore about the predictive values or otherwise of these DNAs, that's for interested parties to research. However what I will comment on is the correlation between Racing codes and I dare say any sport (further testing)? A 92.6% correlation between AUS Gallops-Harness and AUS Greyhounds. In other words build a system and it'll be universal across the two codes. Sample size 16,500 price movements, for AUS Wednesday races, which are somewhat lightly traded (compared to UK), and the money comes very late, in the last few minutes.
One thing pertinent though is that despite a completely distinctly different form of racing, whatever forces move price are near identical in the patterns (DNA) they exhibit, at least for this sample! For some that might be a given, for others though, they're aware of the different dynamics between the two forms of racing, some staying clear of one or the other. Something to ponder?
DNA of last three Matched Betfair Prices
-
rinconpaul
- Posts: 112
- Joined: Wed Dec 03, 2014 10:39 pm
You do not have the required permissions to view the files attached to this post.
-
rinconpaul
- Posts: 112
- Joined: Wed Dec 03, 2014 10:39 pm
Euler,
What's called a unique filter at 2 secs. Every time Volume changes for a runner, but last matched price doesn't = "0". Every time Back or Lay bid changes but Volume doesn't change = "0". Every time Volume changes and matched price differs from last matched price = "1" or "-1".
The trinary value is similar to a Japanese candle, but advises what happened in the last three 2 sec periods.
What's called a unique filter at 2 secs. Every time Volume changes for a runner, but last matched price doesn't = "0". Every time Back or Lay bid changes but Volume doesn't change = "0". Every time Volume changes and matched price differs from last matched price = "1" or "-1".
The trinary value is similar to a Japanese candle, but advises what happened in the last three 2 sec periods.
Hi.
I saw this last week, and didnt have time to think about it .. and am a bit surprised, there isnt more debate.
I think DNA of LTP (or unique filter), is a fantastic, original idea .. and would love to test its application, across sports.
You probably need a fair amount of data, to test etc, but really appreciate the idea, and thx for sharing.
I saw this last week, and didnt have time to think about it .. and am a bit surprised, there isnt more debate.
I think DNA of LTP (or unique filter), is a fantastic, original idea .. and would love to test its application, across sports.
You probably need a fair amount of data, to test etc, but really appreciate the idea, and thx for sharing.
Just my thoughts..
I did take a look at this but nothing jumped out to me.
If you look at 1,1,1 and -1,-1,-1, they are what i would expect anyway (even if you take cross matching into consideration, which is important)?
If you applied it to inrunning and looked at each winner, you would find a very high correlation to -1,-1,-1 and visa versa on the loser..in other words it is likely to be a lagging indicator rather than a leading one?
Regards
Peter
I did take a look at this but nothing jumped out to me.
If you look at 1,1,1 and -1,-1,-1, they are what i would expect anyway (even if you take cross matching into consideration, which is important)?
If you applied it to inrunning and looked at each winner, you would find a very high correlation to -1,-1,-1 and visa versa on the loser..in other words it is likely to be a lagging indicator rather than a leading one?
Regards
Peter