Horse Racing Strategies Megathread
-
@martin with you back the favorite strategy you just made on the latest YouTube video using the favorite filter and then putting bets in at 10 mins before off. Doesn't that give you inaccurate results as the favorite could have changed in the 10 mins till off as the favorite filter is based on BSP.
I been testing your filter on BFBM and noticed a few different selections and was due to changing favorites over that time so was getting a worse results then the filter software says.
Be good if there was a favorite filter that uses the current favorite at the entry time you input
-
@martin said in ***Horse Racing Shared Strategies Megathread + Downloads***:
@keith depends on the strategy I think a lot will be down a bit as the bsp has corrected I think
There's been no change to the BSP figures. The profit could change if you choose a different bet type in the new dropdown, or apply a max number of qualifiers per race, or if the strategy uses the going rule.
-
Short Inrunning looks like a non starter now. Zero Paced Lays ok as are Slow Horses(reduced profit on that one).
The short priced lays still as good. Think it depends on the type of trade it was.
-
@keith depends on the strategy I think a lot will be down a bit as the bsp has corrected I think
-
@john-folan said in ***Horse Racing Shared Strategies Megathread + Downloads***:
@martin said in ***Horse Racing Shared Strategies Megathread + Downloads***:
I'll get some new figures for the strategies this week as I think the update has corrected some of the profit back testing figures, some have changed drastically!
Massively.
@Martin @John-Folan for the better or worse?
-
@martin said in ***Horse Racing Shared Strategies Megathread + Downloads***:
I'll get some new figures for the strategies this week as I think the update has corrected some of the profit back testing figures, some have changed drastically!
Massively.
-
-
-
0_1653081713896_rules_export (6).json
So this is the strategy. The last 4 results were not given to me as selections.
-
Hey lads here’s my strategy “Quick back up lays” initially was looking at a back system as a 7 day back up is normally considered a good thing in aussie racing but flipped that to lay system quickly. Shows the power of back testing. Hoping to get a second set of eyes over my strategy to make sure I haven’t made a mistake before I start forward testing?
0_1653037326246_rules_export.json -
@martin said in ***Horse Racing Shared Strategies Megathread + Downloads***:
@wayne-ironside said in ***Horse Racing Shared Strategies Megathread + Downloads***:
I'm forward testing my strategies and I am finding that some horses that were not on my list are showing up in the results when I run the strategy the next day. I thinking it must be a change in the going of the track as it may have changed from when I downloaded the selections.
When does everyone download their selections? I have been doing in the night before, but I guess I should do it in the morning.
what strategy is this for? can you share the file so we can take a look
Yeah i'll sort it tomorrow been at work all day.
-
@adam said in ***Horse Racing Shared Strategies Megathread + Downloads***:
We've had a couple of tickets about trades appearing in results when they shouldn't, and vice versa.
There are two reasons I've found (stick with me here because there's a question at the end!):
The strategy uses a "Going" rule. You're expecting the going of the race to match the official going on RacingPost. We were trying to be clever here and calculate a more accurate going value based on the RP going allowance. Here is a very old, but still valid, discussion of why you might want to do that: https://www.theukbettingforum.co.uk/XenForo/threads/going-corrections.365/
The upshot is that our going for historical races doesn't necessarily match the official going. The way we're calculating it definitely give a more "accurate" going, but unfortunately we can't know what it is before the race. So if we're trying to be realistic with the backtesting simulator, you'd have to go by the official going, which we know in advance. For this reason, we're going to change this and stop trying to be so clever :face_with_tears_of_joy:
Sometimes people are using an entry price range of, say, 3.1 - 3.1, because they want to lay at exactly 3.1. The problem is, setting your range to that means the horse has to have traded at exactly 3.1. If the actual prices traded were either side of that number, it won't be picked up.
You could change your price range to 1.01-3.1 to pick up these trades, but then your entry price will show as a lower price. For example, if the horse dipped from 3.5 to 2.7, your entry price would be 2.7 and not 3.1.
This isn't realistic if you queue your bets up at 3.1, because the bet would just sit there waiting to be matched and would be matched at exactly that price.
The reason we introduced the entry price range was because sometimes when you're trading, you might just look at the market and say "OK, this horse is now between x and y, I'll open a trade at whatever the current price is".
I think we might need to introduce another option to handle these two different situations. Maybe a choice between a price range, which behaves exactly like now, or a fixed value, which would imply that you queue your bets up at that exact price and wait for them to be matched.
Any other better ideas?
-
Why not have both? Predicted/official going and actual going? Two different options. I have this problem with Bfbot as they use the predicted which can change so I usually check the going after the first race at that meet and dump it if it’s off of what I want it to be.
-
This works lovely and is exactly what I would like. You would need to assume tho that if it trades lower it has traded at that price. Silly but then takes care of your problem.
-
-
Yeah it definitely seems like 2 distinct situations, so warrants a new option. The only question is what to call the option!
Basically you'll choose an option and that will either let you:
- Put in a price range (which is like taking odds anywhere in that range as soon as the horse trades within that range)
- Enter a single, specific price, which is like queuing the bet up at that price and waiting for it to be matched. If it's a lay bet, it will get matched at that exact price as soon as the horse traded at or below that price, and vice versa for backs
-
We've had a couple of tickets about trades appearing in results when they shouldn't, and vice versa.
There are two reasons I've found (stick with me here because there's a question at the end!):
The strategy uses a "Going" rule. You're expecting the going of the race to match the official going on RacingPost. We were trying to be clever here and calculate a more accurate going value based on the RP going allowance. Here is a very old, but still valid, discussion of why you might want to do that: https://www.theukbettingforum.co.uk/XenForo/threads/going-corrections.365/
The upshot is that our going for historical races doesn't necessarily match the official going. The way we're calculating it definitely give a more "accurate" going, but unfortunately we can't know what it is before the race. So if we're trying to be realistic with the backtesting simulator, you'd have to go by the official going, which we know in advance. For this reason, we're going to change this and stop trying to be so clever :face_with_tears_of_joy:
Sometimes people are using an entry price range of, say, 3.1 - 3.1, because they want to lay at exactly 3.1. The problem is, setting your range to that means the horse has to have traded at exactly 3.1. If the actual prices traded were either side of that number, it won't be picked up.
You could change your price range to 1.01-3.1 to pick up these trades, but then your entry price will show as a lower price. For example, if the horse dipped from 3.5 to 2.7, your entry price would be 2.7 and not 3.1.
This isn't realistic if you queue your bets up at 3.1, because the bet would just sit there waiting to be matched and would be matched at exactly that price.
The reason we introduced the entry price range was because sometimes when you're trading, you might just look at the market and say "OK, this horse is now between x and y, I'll open a trade at whatever the current price is".
I think we might need to introduce another option to handle these two different situations. Maybe a choice between a price range, which behaves exactly like now, or a fixed value, which would imply that you queue your bets up at that exact price and wait for them to be matched.
Any other better ideas?
-
Sorry for the silly question but having read the last few posts am I right in thinking that if the time filter is set for 1 hour before the race in the case of picking the favourite it will pick the horse with the lowest price at that exact time and record if it wins or loses regardless if it started as favourite or not.
-
@wayne-ironside Had the same thing myself, logged a ticket with some of my specifics. I do them on the morning of the race.
-
@svajunas-bernotas-0 said in ***Horse Racing Shared Strategies Megathread + Downloads***:
this one gives 4.2K profit on £1 stakes at market open price
but if filtered by entry time it only shows 1 bet 1-2 hrs before race and 13k bets at market open
I get the same number of bets if all I change is the entry time to 1 hour before the race. The reason the profit is different is because you'll be taking a different price on each runner at the opening time of the market vs 1 hours before the race.
