The BTC Horse Racing Thread
-
The Beaten Favourite strategy has been a disaster this month, judging by the back testing this month is an outlier. Personally I think it's a case of leaving it as is, and let it run until there is enough data to suggest that the strategy needs tweaking. There is always going to be variance, it's just not much fun when you are going through a drawdown.
-
I never switch to paper trading when a new system hits a bad run, I carry on. Most of the time they pick up if the research is good. If it breaks the bank, fine. I then do further research, and start again. In most cases, with an additional filter/lower stake. No use binning something you had faith in to starts with.
As for season changeovers, that happens in April/October. It's just swings and roundabouts at other times.
-
@George-Bowling yeah, the only thing I would consider without doing a real deep dive or separating the strategy out into variations is to have a type of stop loss and paper trade.
For example if you had a 100pt bank and had a 50pt stop loss and 65pt to restart. When your bank hits 50pt you switch to paper trading until it recovers to 65+ and go back to real trading. Or if not points you could try with time, paper trade 2 weeks or 1 month, etc.
-
@Dan-MacKinnon thats interesting, thanks.
I have had similar thoughts on the change in seasons between races, it is something I have picked up on when running other strategies in the past and things like post Cheltenham dips etc.
I had already been looking if DSLR had much of a factor in this and if a beaten favourite was last run at the end of last season if this plays a factor in some of the heavier loosers we have seen this month. Some of the filters I tried certainly reduced the loss of this month, but subsequently, some of the previous months too seem to not be as profitable/some more months being loosers overall too.Will keep playing with it but nothing immediately obvious. Hopefully settle down over summer flat season.
-
@George-Bowling I've been trading it since around November of last year. I think every month has been break even or profitable except for December and now this month.
This month has been horrendous, and not sure if that's potentially due to the changeover from winter to summer races. I can't give an accurate assessment because I've been messing about with bots and withdrawing funds but it needs a strong summer to bring it back on course.
-
@Matthew-Harrison Been a rollercoaster month since I started forward testing too.... up over 50pts at one stage and down minus 20pts now on LTBF alone. I have faith it should settle down again soon though. How long have others been forward testing/ using the LTBF and hows the overall picture looking for people?
-
I’m down almost £100 off £1 stakes in the last couple of weeks from ltbf. I’m going to paper trade until it picks up
-
@Joseph-Henderson yeah very bad
hoping it turns -
Tough spell for the LTBF strategy. Good test of discipline I guess..
-
@Joseph-Henderson risk Vs reward so depends what the max liability is. When I get it where I want I will leave it on a 200 point bank. Nice and safe then.
-
@John-Folan said in The BTC Horse Racing Thread:
@Joseph-Henderson I compound at various levels depending on drawdown until I reach the max stake I want. Compounding via bot after each trade is the most efficient way to build a bank. Anyone who tells you otherwise is leading you up the garden path. It means there’s no nasty shocks if you have a downturn like you would if you did the old double your bank then double your stakes route. Much better for the old blood pressure.
How do you tend to decide the max stake you are comfortable with for a strategy? And how do you change your staking plan once you get to that desired stake?
-
@Dan-MacKinnon I would have agreed with you up until last week, in fact I had it set up that way and was very much enjoying the compounding of my bank at breakneck speed! The problem is that your losses get exaggerated especially when there are a few in a row. For example, my bank was cut in half by those three or four losses last week staking at 1% with the ratchet enabled. I then took the ratchet off and a further loss only cost me 7% of my bank rather than another 15-20% that would have gone with ratchet enabled.
Each to their own, as your say if I had maybe been staking 0.5% it would have been manageable but I quite like that the losses stay in sync with the overall bank.
-
@David-Milligan when is ratchet staking useful? I think it's good for LTBF because the losing runs are quite low. However, when they do happen it's good to keep the same stakes to recover quicker assuming you've staked appropriately in the first place
-
@Joseph-Henderson Yea, it's a bit too aggressive I think. With a strategy like this where the S/R is high there's no need for it, especially with the liabilities getting high when flat staking. Not sure what situations would be a good fit for it.
-
@David-Milligan said in The BTC Horse Racing Thread:
@Joseph-Henderson I had these settings too for LBF but found I was staking up to 2% or more of bank after a few losses so I've removed the ratchet
Just been pondering, when would enabling ratchet be beneficial? Surely this would always be the case when you hit a bad run..
-
@John-Folan 100%… I have generated the average odds for wins and losses for all the strategies I use for the last two years and complied them on a monthly breakdown.. it’s been amazingly helpful to track my actual results as I go forward against corresponding months from the last couple of years. It’s really helpful to know if a good/bad run is an outlier or just part of the random distribution. When the average odds broadly match those on the back testing, I feel much more secure when there is a rough patch..
-
@John-Folan said in The BTC Horse Racing Thread:
The other thing I find incredibly useful is to keep an average odds figure. That way I can keep an eye on the probability to make sure it’s long term profitable.
This is in the next update which I will probably release tomorrow
-
The other thing I find incredibly useful is to keep an average odds figure. That way I can keep an eye on the probability to make sure it’s long term profitable.
Read the first post on both threads.
Horse Trading Resources
https://forum.betfairtradingcommunity.com/topic/1111/john-s-horse-trading-resourcesBfBotmanager
https://forum.betfairtradingcommunity.com/topic/2915/bfbotmanager-automation-discussion -
@Adam yes. I’ll send you an Asana. I think I tried to explain it before but I probably sounded like I was talking French.
Basically max drawdown expressed as a points value being the lowest point dropped to after a previous high.
So strategy hits 160 points as a high then drops 6 points, recovers 2 but then drops another 4 would have a max drawdown of 8 points.
Then after that it hits a high of 200. Drops 10, recovers 5 but drops another 10 to 185 would have a max drawdown of 15 points.
Probably a good idea to express it as a percentage as well as number of points. I find both really useful. Especially as a big number of points dropped after 6 months of running a strategy is a lot lower than if you had had the same drop after 1 week as (for example) a fifty point drop (with little recoveries and further drops)immediately is 50% of your 100 point bank but later on would only be 25% of your 200 point bank as the size increases if it’s a good strategy.
I’d never thought of using the percentage of bankroll as a max drawdown but I really like using it as an extra tool now.
