Backing purely on those stats probably would not work long term - the odds will already factor that in, its a little too simplistic. The key is finding reasons those greyhounds WONT be the true favourite that are not obvious. Though I think that may already be what you are suggesting as I can see on the bookies that the greyhound 'long shot' is the outsider. Is your list of backs based on none favourites that are showing better than expected stats?Atho55 wrote: ↑Sun May 05, 2024 8:36 amThese have been derived from Memphis`s latest data. Have tried to find the fastest runners at Track/Dist/Grade when you combine those 3 bits of info into a single text string all based upon the last race. It`s a list of runner names only unfortunately. Potential BACK candidates.
0505 Back.png
I ran my own prediction model today, and filtered to dogs that all ran at the same grade in their previous race in an attempt to simplify the variables that may increase error in my model.
From my filtered list, I found from your list Long Shot is ranked 6th with my model. Considering Memphis input, your list, and my model I would be looking to trade the dog ''long shot'' (how ironic). I use the rank of dogs from my model rather than the predicted odds to inform my trading as I find the odds can be a little wild at times.
I'm currently trying to think of a way to collaborate with Memphis in a mutually beneficial way but have not had the time to really test my model over a large sample and then tie this in with what Memphis produces. I've considered sharing my model predictions daily and letting the members figure out strategies. But the problem is that A) my output is not the most user friendly in its current format and B) That may create user dependency should I decide I no longer wish to share my output.
Attached are my full predictions and filtered list for the day. Personally would recommend the filtered list as my full raw list is a bit of a pain to manage and doesn't like it when there are time trials as previous grade and a few other random bugs with the odds.
Heres a snippet from my sheet for the race with long shot in. Name, grade of previous race, predicted true odds (big pinch of salt) and rank based on odds. Bookmakers currently have Long shot at odds 11.0 so may be some trading potential.
SWIFT GIDDY 'OR3' 3.75 Rank 1
LONG SHOT 'OR3' 14.55 Rank 6
BLUE RANGE 'OR3' 3.99 Rank 3
RAZLDAZL BARKLEY 'OR3' 14.16 Rank 5
HEATHLAWN WILLOW'OR3' 12.01 Rank 4
MAKEIT MAGICAL 'OR3' 3.84 Rank 2
