Memphis, haven’t we already established that a dogs, horses, golfer, tennis player, snooker player, tarts player or any other sports professionals previous form has no bearing on the race or game being played or did I miss something over the years ?MemphisFlash wrote: ↑Wed Apr 17, 2024 7:47 pmAll myself, theres not a lot of code to it really,
a bit of recording macro's then manipulation from there
Today's Greyhounds
-
- Posts: 3298
- Joined: Thu Oct 24, 2019 8:25 am
- Location: Newport
- MemphisFlash
- Posts: 2222
- Joined: Fri May 16, 2014 10:12 pm
- Location: Leicester
yep, that is right, i just got bored not interacting with my bot,
so decided to be creative,
hard sittting on my hands doing nothing
so decided to be creative,
hard sittting on my hands doing nothing
-
- Posts: 3298
- Joined: Thu Oct 24, 2019 8:25 am
- Location: Newport
MemphisFlash wrote: ↑Wed Apr 17, 2024 8:48 pmyep, that is right, i just got bored not interacting with my bot,
so decided to be creative,
hard sittting on my hands doing nothing
i heard that weemac's garden needs a few hours weeding!? (and by your own admission, your own garden )MemphisFlash wrote: ↑Wed Apr 17, 2024 8:48 pmyep, that is right, i just got bored not interacting with my bot,
so decided to be creative,
hard sittting on my hands doing nothing
- MemphisFlash
- Posts: 2222
- Joined: Fri May 16, 2014 10:12 pm
- Location: Leicester
Another twist on greyhound selections,
has anyone ever considered laying each dog that has the worst trainer at each track.
I hadn't but figured how to do it.
has anyone ever considered laying each dog that has the worst trainer at each track.
I hadn't but figured how to do it.
You do not have the required permissions to view the files attached to this post.
I heard a story many years ago about a Trainer that over fed his dogs, he would run 2 of his best dogs in the same race knowing that they would be lethargic from the food and bet on the others. Maybe you need to take this and probably many other goings on in dog racing into account. Look out for dogs pooing as they may have been over fed. But the Automation cant take that into account.MemphisFlash wrote: ↑Fri Apr 19, 2024 1:16 pmAnother twist on greyhound selections,
has anyone ever considered laying each dog that has the worst trainer at each track.
I hadn't but figured how to do it.
Capture.PNG
Capture.PNG
This has been happening for years in the UK, the ruling bodies in the last century in Greyhound racing were well aware of this and tried to combat it by putting in limits on the weight a greyhound could put on between races it currently stands at 1kg, I believe.gstar1975 wrote: ↑Fri Apr 19, 2024 3:03 pmI heard a story many years ago about a Trainer that over fed his dogs, he would run 2 of his best dogs in the same race knowing that they would be lethargic from the food and bet on the others. Maybe you need to take this and probably many other goings on in dog racing into account. Look out for dogs pooing as they may have been over fed. But the Automation cant take that into account.MemphisFlash wrote: ↑Fri Apr 19, 2024 1:16 pmAnother twist on greyhound selections,
has anyone ever considered laying each dog that has the worst trainer at each track.
I hadn't but figured how to do it.
Capture.PNG
Capture.PNG
Trainers are well aware of this but also know a dogs performance can alter significantly at 0.5kg. The only trouble is racing weights as far as I know are still only published at the stadium in the UK. Knowing a dog is at it's perfect weight of its previous race wins is a key piece of knowledge.
- wearthefoxhat
- Posts: 3243
- Joined: Sun Feb 18, 2018 9:55 am
Avoid the runners dropped in grade/class (similar to horse racing) as the trainer's win strike rate will then improve when they start winning again.MemphisFlash wrote: ↑Fri Apr 19, 2024 1:16 pmAnother twist on greyhound selections,
has anyone ever considered laying each dog that has the worst trainer at each track.
I hadn't but figured how to do it.
Capture.PNG
Capture.PNG
-
- Posts: 3298
- Joined: Thu Oct 24, 2019 8:25 am
- Location: Newport
And, as we know, avoid those dogs coming across from Ireland and usually put into traps 3 or 4!!!!!!wearthefoxhat wrote: ↑Fri Apr 19, 2024 5:54 pmAvoid the runners dropped in grade/class (similar to horse racing) as the trainer's win strike rate will then improve when they start winning again.MemphisFlash wrote: ↑Fri Apr 19, 2024 1:16 pmAnother twist on greyhound selections,
has anyone ever considered laying each dog that has the worst trainer at each track.
I hadn't but figured how to do it.
Capture.PNG
Capture.PNG
Been messing with my prediction model and well... its bloody good. I will keep posting its data if people find it useful. I haven't yet figured out the best way to utilise this - its a lot of manual trading so would be interested if someone finds a good way to automate profit from it as the profits are good but its just so tedious manually trading. I may be willing to provide the prediction model to anyone that can create a profitable automation for it. So far I've found trading with less than 3 mins to go is a good entry. I've attached the predictions for tomorrows races. Annoyingly my computer disconnected with about 50 dogs still to harvest so there will be a few races missing.
How to use:
Ctrl + F , enter a dogs name from the race you are looking at
Do not take the odds literally, they are wrong. I repeat, do not take the odds literally. Use the odds to spot false favourites. The ordering of dogs is pretty good. It is better to lay a heavy favourite that has another dog or ideally 2 dogs shorter priced from the model. If you see the model is predicting odds 9.0 and there is a dog at 4.0, dont assume its going to lengthen, thats not how ive been using the predictions. Use the odds as a guide to the order the dogs should be, especially at the favourite end of the market.
Grd - This is the grade of the dogs last race. The model does not consider changes in grade. So dogs that previously ran at lower grades will likely perform better than predicted as they are dragged along by quicker dogs. Its probability calculations come from times and some secret x factors.
If a race contains some strange values like negative odds, may be worth just ignoring that race. Seems to be some small bugs in the model.
Time trials - if a dog ran a time trial under 'grade' this frequently messes with the odds and shortens the odds drastically. Again beware of this.
How to use:
Ctrl + F , enter a dogs name from the race you are looking at
Do not take the odds literally, they are wrong. I repeat, do not take the odds literally. Use the odds to spot false favourites. The ordering of dogs is pretty good. It is better to lay a heavy favourite that has another dog or ideally 2 dogs shorter priced from the model. If you see the model is predicting odds 9.0 and there is a dog at 4.0, dont assume its going to lengthen, thats not how ive been using the predictions. Use the odds as a guide to the order the dogs should be, especially at the favourite end of the market.
Grd - This is the grade of the dogs last race. The model does not consider changes in grade. So dogs that previously ran at lower grades will likely perform better than predicted as they are dragged along by quicker dogs. Its probability calculations come from times and some secret x factors.
If a race contains some strange values like negative odds, may be worth just ignoring that race. Seems to be some small bugs in the model.
Time trials - if a dog ran a time trial under 'grade' this frequently messes with the odds and shortens the odds drastically. Again beware of this.
You do not have the required permissions to view the files attached to this post.
Last edited by Fugazi on Sat Apr 20, 2024 9:20 am, edited 1 time in total.
-
- Posts: 1088
- Joined: Fri Nov 20, 2015 9:38 am
That made me smile....
I haven't had a look at but as far as automation is concerned have you considered just looking for "value" opportunities based on results of your model - if anything that should be easier to automate. And there is no need to do every selection, just pick best or worst & back (or lay) at what you expect to be a value price & see how it goes?
edit - and don't worry if you don't always get match, it just means you are probably in the "value" range
edit edit - but do worry if you always get matched (as probably need more value)!
Maybe Im biased hah.sionascaig wrote: ↑Sat Apr 20, 2024 7:49 amThat made me smile....
I haven't had a look at but as far as automation is concerned have you considered just looking for "value" opportunities based on results of your model - if anything that should be easier to automate. And there is no need to do every selection, just pick best or worst & back (or lay) at what you expect to be a value price & see how it goes?
edit - and don't worry if you don't always get match, it just means you are probably in the "value" range
edit edit - but do worry if you always get matched (as probably need more value)!
Yeah Ive been trading so many races it burns me out. I've been wondering if simply finding the best picks and laying at BSP is the easiest way. The best picks being where there is a massive favourite in the betting yet my model is predicting another 2 dogs at very close odds to the favourite.
What I'm unsure of is whether Im profiting because im entering and exiting the market or the right time, or if its actually possible to profit at BSP which will make automation easier. Maybe something on the lines of Lay dog 1 at BSP if it is not dog 4 . I think I can get it to import dog names from a spreadsheet or something rather than manually set up a new automation file for every race. Maybe need to stick my head back in the instruction manual.
The most common way trainers get a good gamble on their dog is to let a fast dog get downgraded after a loss. So if a dog can run close to 29 seconds in a A2 grade race over 500m, trainers will overtrain a dog prior to a race so it's tired. They don't need to overfeed it. Just tire it out prior to kenneling. It runs 32 seconds or more. Bingo.
Next race out, he optimizes the dog and it beats the lower A3 grade field by several lengths.
Next race out, he optimizes the dog and it beats the lower A3 grade field by several lengths.