If you are betting on short priced favourites between 1.4 and 1.8 and you find youself on a winning streak, at what point can you conclude that your winning streak is down to an edge rather than random chance?
As the circumstances I need only happens 3 or 4 times a day it is taking quite a while to test my theory but at the moment I have 8 wins on the "trot" and no losses.
Can this be luck or should I put the champagne on ice?
What are the chances of that?
Gazuty
Thanks for that.
I saw that link before my holidays and intended checking it out..
What did you use for the Prob of a single trial?
For EG; Assume you were only laying and only ever offered a price at reverse odds. Further assume that there was always a gap between prices
EG Lay at 2 (Current Lay price = 2.3)
Or would you use the strike rate: EG I was sucessfull x% of the time? Eg Sucessfull 1 in 3 races, ie 0.333
Thanks
regards
Peter
Thanks for that.
I saw that link before my holidays and intended checking it out..
What did you use for the Prob of a single trial?
For EG; Assume you were only laying and only ever offered a price at reverse odds. Further assume that there was always a gap between prices
EG Lay at 2 (Current Lay price = 2.3)
Or would you use the strike rate: EG I was sucessfull x% of the time? Eg Sucessfull 1 in 3 races, ie 0.333
Thanks
regards
Peter
(Editing.) If you have done enough trials of a strategy - strike rate of the strategy.PeterLe wrote:Gazuty
Thanks for that.
I saw that link before my holidays and intended checking it out..
What did you use for the Prob of a single trial?
For EG; Assume you were only laying and only ever offered a price at reverse odds. Further assume that there was always a gap between prices
EG Lay at 2 (Current Lay price = 2.3)
Or would you use the strike rate: EG I was sucessfull x% of the time? Eg Sucessfull 1 in 3 races, ie 0.333
Thanks
regards
Peter
If you were doing a single lay at 2 in an efficient market (back/lay book @ 110.1/99.9) then I'd say the probability is 0.5.
Ok. Just for arguments sake let's say you are backing at an average of 1.6.guy333 wrote:If you are betting on short priced favourites between 1.4 and 1.8 and you find youself on a winning streak, at what point can you conclude that your winning streak is down to an edge rather than random chance?
As the circumstances I need only happens 3 or 4 times a day it is taking quite a while to test my theory but at the moment I have 8 wins on the "trot" and no losses.
Can this be luck or should I put the champagne on ice?
Efficient market says your probability of winning is 0.625.
8 trials @ 0.625 for 8 successes. Probability = 0.023 ie 2.3%
Not small enough to prove an edge.
Come back once you've done 100 trials.
What would we expect? 63 wins and 37 losses. Let's be modest.
£10 stakes. Wins 63 X £6 X 0.95 = 359.1. Losses 37 X £10 = 370
Expected outcome -£10.90
What would the probability be in 100 events of getting 70 or more wins? 7.2%
Scale up to 1000 events
What would be the probability of getting 700 or more wins @ 0.625 probability? 0.0001%
So if you can do that on 1000 events you have an edge (and my respect) and you can uncork the champagne.
Ok thanksgazuty wrote:(Editing.) If you have done enough trials of a strategy - strike rate of the strategy.PeterLe wrote:Gazuty
Thanks for that.
I saw that link before my holidays and intended checking it out..
What did you use for the Prob of a single trial?
For EG; Assume you were only laying and only ever offered a price at reverse odds. Further assume that there was always a gap between prices
EG Lay at 2 (Current Lay price = 2.3)
Or would you use the strike rate: EG I was sucessfull x% of the time? Eg Sucessfull 1 in 3 races, ie 0.333
Thanks
regards
Peter
If you were doing a single lay at 2 in an efficient market (back/lay book @ 110.1/99.9) then I'd say the probability is 0.5.
Number of Trials: 9674
Successes : 1747
Strike : 18.06
Therefore Cumulative Probability : 0.505
I wasnt going to post this but I thought it worthwhile ilustrating the graph. (it also fits in the general theme of Peter's rcent post)...Might help someone new to this type of thing..?
This strategy was based on a hunch. You can see from the graph that it didnt fair too well initially (but I was almost certain it would work long term).
After I had a relative amount of data, I extracted the bet history to Excel, and created a Pivot to see what was working and what wasnt. I removed the negative going forward. After than there was an initial surge in profits, but then it started to stall.
Rinse Repeat the data Extraction/Pivot etc
Im fairly sure this will work long term, so much so, I've now started to compund the profits by increasing the stakes by a tiny fraction.
Projections based on compounding are frigtening! Been here many times before, so we will have to see!
Anyway thanks for input Gazuty
Regards
Peter
You do not have the required permissions to view the files attached to this post.
Meanwhile back on the original question.
It seems that it doesn't matter how successful a system is in practice mode it will inevitably fail when the real money enters the stage.
Got to 18 wins out of 21 races which I thought was a reasonable success rate but as soon as I went into live mode I lost 2 in a row.
What are the chances of that ????
Still 18 out of 23 is still pretty good so I am not going to abandon ship straightaway, so frustrating though.
It seems that it doesn't matter how successful a system is in practice mode it will inevitably fail when the real money enters the stage.
Got to 18 wins out of 21 races which I thought was a reasonable success rate but as soon as I went into live mode I lost 2 in a row.
What are the chances of that ????
Still 18 out of 23 is still pretty good so I am not going to abandon ship straightaway, so frustrating though.
21 races is nothing, you need at least couple of thousand spread across different seasons to make any conclusionsguy333 wrote: Got to 18 wins out of 21 races which I thought was a reasonable success rate but as soon as I went into live mode I lost 2 in a row.