The BTC Horse Racing Thread
-
@dan-mackinnon said in The BTC Horse Racing Thread:
@nick-allan I've only recently got into horse racing but looking at those stats how many traders would love a strategy which only has 1 losing month in 18?! If you were doing flat stakes that month cost you 1% of your bank. As John has mentioned before, the software isn't perfect but even if we allowed for 10% error it's still decent.
I know there were a couple of comments on the football thread about focusing on one strategy at a time. With the horse racing software, I find it easier to focus on one strategy and use it as a guideline. There will be some differences but if you're roughly in line with the software then happy days!
I'm sure the thought of laying at up to 10/1 and hitting a couple of losers is pretty scary for most of us. So I thought I'd try running the Pace Lays with Dan's revisions but with a £1 liability rather than stake. It gave me a profit of £61.81 and ROI of 4%. Not massive but it does mean the high price losers won't hit the bank so hard...
-
@nick-allan I've only recently got into horse racing but looking at those stats how many traders would love a strategy which only has 1 losing month in 18?! If you were doing flat stakes that month cost you 1% of your bank. As John has mentioned before, the software isn't perfect but even if we allowed for 10% error it's still decent.
I know there were a couple of comments on the football thread about focusing on one strategy at a time. With the horse racing software, I find it easier to focus on one strategy and use it as a guideline. There will be some differences but if you're roughly in line with the software then happy days!
-
@dan-mackinnon said in The BTC Horse Racing Thread:
After watching Martin's video of his Pace Lays strategy I went through it in the software. It looks like a decent strategy with £299 profit from flat stakes and only 2 losing months (May 22 -10.55 and Jan 22 -4.34) the best month is Sept 22 with 45.14 and the next 5 are between 27-33.
I made 2 minor adjustment which was to remove Class 1 races and change the start time from "actual" to "official". This took the overall profit up to £329.95, a 10% increase. It also reduced the longest losing run from 4 to 3. It changes May from being the worst performing month to being 3.95 in profit, and January is now the only losing month -1.01. It does bring the most profitable month down from 45.14 to 42.22 and only 4 months between 27-33, however it's a better P/L over all months.
In addition to this, I broke down each month to get an idea of how it'll feel to trade this day to day. I've put the record below
As you can see, even though there's a 86% SR there are some pretty rough losing days, but no outstanding profitable days. If you follow this strategy you have to be prepared that most months you will have a day that loses 10+ pts. On 4 occasions you'll have 3 days in a row that lose and one of them could well be 10-15% of your bank. I know on the football thread when we have a bad weekend it's easy to feel defeated, this way you should be mentally prepared. Also, it shows you don't have to get on loads of trades, there are only 2 months which are 100 plus.
Finally, I think it's always good to run strategies both as flat stakes and compounded. This shows what would seem like a profitable strategy with flat stakes soon becomes a disaster when you start adjusting them. For this I used £100 bank and 1% stakes, the overall profit is £2,246. Jan 22 still remains the only losing month with -14.42 now, but Sept 22 is 593.08 and the next 4 months are all 100+
I just wanted to share this as a couple of people have asked about the horse racing software. It won't magic a winning strategy for you, but helps you prepare for the rollercoaster! I never truly appreciated how great the software is for doing that and how good trading horses can be
Some great analysis dan, ive been testing this to £1 stakes and started the beginning of october so only have one month of results and that was £22.36 so +22.36 which is very impressive, but yes the dreaded 11.0 lay loss hasnt come but that will be a bank wrecker and if you get 2 in a day its devestating. I think the worse i had in oct was 6.8. V intresting about taking out class 1 races, perhaps it means horse and jockeys perform better in C1 and you maybe subjected to a big loss but definately worth taking all C1 races out of the filter on that evidence. Good work buddy
-
@john-folan said in The BTC Horse Racing Thread:
Then Compounding at 1% of a 100 point bank. This between the two would give you over 5k in 18 months. Trading bank built. Feel free to download and mess with it. DSLR filters, probably not a bad place to start. I haven't looked. Not bad work for a Sunday morning and two cups of coffee...
Back The Favourite at 1% MTSa
0_1667125642923_BTFMTSa.json
Lay the Favourite at 1% WThFSu
0_1667125814052_LTFWTFSu.json
wow!!! some very intresting stuff john, ive stopped the shorts odds lays on saturday as both yourself and others have come up with some compellling arguements backed up with great data. Think ill add this back the fav to my testing strategies as i like it and i have a bot ready to go. I take it this back the fav MTS is at any odds? i noticed below that results for 4 - 4.9 was poor which was odd, so just wanted to check that it would be backing at any BSP odds?
-
Is there a way to check how much you would be staking on each trade? I can see the entry/ exit price and the P/L
Following on from the below I wanted to check that there wasn't unrealistic staking skewing the results. It looks like Sept 22 was around £25 stakes
-
After watching Martin's video of his Pace Lays strategy I went through it in the software. It looks like a decent strategy with £299 profit from flat stakes and only 2 losing months (May 22 -10.55 and Jan 22 -4.34) the best month is Sept 22 with 45.14 and the next 5 are between 27-33.
I made 2 minor adjustment which was to remove Class 1 races and change the start time from "actual" to "official". This took the overall profit up to £329.95, a 10% increase. It also reduced the longest losing run from 4 to 3. It changes May from being the worst performing month to being 3.95 in profit, and January is now the only losing month -1.01. It does bring the most profitable month down from 45.14 to 42.22 and only 4 months between 27-33, however it's a better P/L over all months.
In addition to this, I broke down each month to get an idea of how it'll feel to trade this day to day. I've put the record below
As you can see, even though there's a 86% SR there are some pretty rough losing days, but no outstanding profitable days. If you follow this strategy you have to be prepared that most months you will have a day that loses 10+ pts. On 4 occasions you'll have 3 days in a row that lose and one of them could well be 10-15% of your bank. I know on the football thread when we have a bad weekend it's easy to feel defeated, this way you should be mentally prepared. Also, it shows you don't have to get on loads of trades, there are only 2 months which are 100 plus.
Finally, I think it's always good to run strategies both as flat stakes and compounded. This shows what would seem like a profitable strategy with flat stakes soon becomes a disaster when you start adjusting them. For this I used £100 bank and 1% stakes, the overall profit is £2,246. Jan 22 still remains the only losing month with -14.42 now, but Sept 22 is 593.08 and the next 4 months are all 100+
I just wanted to share this as a couple of people have asked about the horse racing software. It won't magic a winning strategy for you, but helps you prepare for the rollercoaster! I never truly appreciated how great the software is for doing that and how good trading horses can be
-
I always treat software as a guide. If I am getting as near as dammit to the software I am happy. There will always be the odd difference. Some will work in my favour and some won’t. It will even itself out. Preferably while I am wallowing in a bath of money. There’s a nice image for you
-
@john-folan said in The BTC Horse Racing Thread:
Good start. Backs tomorrow
Backs did ok as well. Two points to the good. Nice to get off to a winning start. So three points up across the two days. The forward testing has begun.
One point to note is I have set this for one bet on each horse but if the favourite changes after my back/lay goes in it will also trade that horse as well. I’m trading 10 seconds out as I will beat bsp more often than not. Close enough the favourite won’t change that often.
-
Good start. Backs tomorrow
-
@martin said in The BTC Horse Racing Thread:
@john-folan flip me mate I might have to put some time aside this week to go through those posts fully! Good work!
Enjoy. I have wittered on a bit. Just to show how my strange mind works. Last two posts are probably the best
-
@john-folan flip me mate I might have to put some time aside this week to go through those posts fully! Good work!
-
@dan-mackinnon said in The BTC Horse Racing Thread:
This might be a daft question, but what is "Max. Drawdown"?
For example, the Back The Favourite below has a drawdown of 22%. Am I right in thinking that's how much of your bank you could potentially lose?
Exactly that.
-
This might be a daft question, but what is "Max. Drawdown"?
For example, the Back The Favourite below has a drawdown of 22%. Am I right in thinking that's how much of your bank you could potentially lose?
-
@kevin-pepper said in The BTC Horse Racing Thread:
@John-Folan Great stuff John. Any thoughts why Mon & Tues seem to be strong days for Faves? I thought I'd have a look on Adrian Massey and just a basic all clear fav check shows them being profitable based on 2017-22 so it's not a recent thing. Saturdays however aren't so healthy and show quite a loss.
That brings me onto my other thought which is whether/when we'll have access to more year's results to be able to check some of the great filters over a longer timescale
@Adam - any update on adding more historical data?
No idea tbh. Could just be the type of racing. (Class?). Not sure on the Saturdays and Massey’s data. He has updated that site once in the last three years manually. The rest of it is an auto update so wouldn’t completely trust it although its a nice double check. I know where Adam gets our data from and it’s straight from the horses mouth so as to speak. Pain in the arse he still uses fractional odds as well. But that’s just me being fussy .
I’m going to set these up on bots with 100 banks and see how they crack on going forwards. Proof is in the pudding and all that.
-
@John-Folan Great stuff John. Any thoughts why Mon & Tues seem to be strong days for Faves? I thought I'd have a look on Adrian Massey and just a basic all clear fav check shows them being profitable based on 2017-22 so it's not a recent thing. Saturdays however aren't so healthy and show quite a loss.
That brings me onto my other thought which is whether/when we'll have access to more year's results to be able to check some of the great filters over a longer timescale
@Adam - any update on adding more historical data?
-
Then Compounding at 1% of a 100 point bank. This between the two would give you over 5k in 18 months. Trading bank built. Feel free to download and mess with it. DSLR filters, probably not a bad place to start. I haven't looked. Not bad work for a Sunday morning and two cups of coffee...
Back The Favourite at 1% MTSa
0_1667125642923_BTFMTSa.json
Lay the Favourite at 1% WThFSu
0_1667125814052_LTFWTFSu.json
-
So here we have to flat £1 stakes with the classification filters added
Backing the Favourite Monday, Tuesday and Saturday
Laying the Favourite Wednesday, Thursday, Friday and Sunday
-
So that is two simple strategies from one idea. The only other thing I will be doing is classification filters. They will make a massive difference.
The Backing Classifications
The Laying Classifications
-
And again. £100 bank 1% compounding and you have £1200 in 18 months. File attached.
0_1667122822263_BTFMWSa.json
-
Class again. I shall be losing the Class Ones.