Is anyone actually profitable live trading?
-
Hello,
New to the BTC, having watched the BTC YouTube channel, specifically for horse racing.
Before proceeding to commence with the 4 core strategies that are listed on here, I wanted to read through the latest forum posts.
From the feel I was getting from the posts, is that whilst the back testing of each of the strategies stack up. When put into live trading, they don’t? Is this right?
I am talking specifically regarding the core “preset” strategies.
Interested to hear people’s feedback.
Thank you
-
Has anyone else found that quite a few strategies seem to have almost flipped round when you compare 2022 with 2023 (nearly a full year of results). Guess the challenge is to crack that and find something that is consistent over the two years of data (I'm ignoring 2021 for now due to covid disruption). Any thoughts?
-
@Adam Cheers for replying on the other thread.
I've got my P value set up the wrong way round in google sheets so 1 is good and 0 is bad (don't ask why)
So the 0.99728 is definitely "good", and so corresponds to p value of 0.00272 if done the correct way around, apologies if that confused anyone
-
@Andrew-Gray said in Is anyone actually profitable live trading?:
Hi Martin,
Can you elaborate on Archie score please?
I looked it up and got the impression that 0.25 means the results are highly likely down to chance and no specific edge on the market... is that correct?Essentially, yes. Here is a table which relates different scores to implied probabilities:
@Adam - Is this something worth looking into?
See my response on the other thread where you have raised the same thing: https://forum.betfairtradingcommunity.com/post/170181
@Andrew-Gray said in Is anyone actually profitable live trading?:
The P value on this one is 0.99728 though, which I am led to believe means the results are highly statistically significant
Generally with p-values, people look for a value lower than 0.05 (although frankly, that number is a bit arbitrary). 0.99728 would be a very high value and indicates a very high probability of the null hypothesis being true, in other words the opposite of what you're hoping for.
-
@Andrew-Gray yep 0.25 Archie score is means it's one to avoid, that's why I'm loving using that metric for forward testing.
-
Also here is my comparison of the BTC back testing and my BFBotmanager bets since 19th September for the Short Odds Lays filter:
EDIT: My figures for the screenshot before were slightly wrong due to some duplicate rows. Was 34pts profit before but now is only 30pts, but the ROI is actually now higher (17% rather than 10.8%)
However as this filter is purely recreated in BFBM and nothing to do with receiving the selections from the BTC link I guess this explains it.
Still all a bit odd and really makes it hard to know whether you are on to something or just being lucky. The P value on this one is 0.99728 though, which I am led to believe means the results are highly statistically significant, so will obviously just keep rolling with this one!!!
I will post up some other analysis of the other strats if I find anything else worth posting
-
Hi Martin,
Can you elaborate on Archie score please?
I looked it up and got the impression that 0.25 means the results are highly likely down to chance and no specific edge on the market... is that correct?@Joseph-Henderson - I have been running alot of the member suggested and other "recommended" strategies and have finally had a bit of a chance to crunch all results into a decent set of spreadsheets to compare the data correctly.
The now not recommended (as far as I am aware anyway) Lay the short price favourites filter is performing the best by far. This is 34pts up since 19th Sept.
Lay 3rd Fav for Turf and AW only is also in profit since 30th October. However I compared that to BTC software for the same dates and this is the comparison:
Little bit odd. I had a brief look over the individual trades (I know there will be a few here or there that don't match up but nearly 50 trades missed seems way off) and found a few weird selections in there, for example this one:
Cannot see how something that was 2.42 can ever have been the 3rd favourite at time of bet (or ever tbh).
@Adam - Is this something worth looking into?
Going back to the general discussion on whether any of this can be long term profitable.
Many of the other strats have been pretty poor or just not doing much recently, as Joseph says, but they have been weirdly profitable up until maybe 2-3 months ago.
This suggests a few different possibilties to me:
- They were just data mining/overfitting and now results are reverting back to the mean or just randomness
- There are so many people running these strategies via BTC that we are moving the price towards the "not value" side of the equation
- The backtesting software is not accurate enough (see my anomaly above for example)
- Whatever trends/edges that have been found in the backtesting software for the last ~2 years have for some reason been eroded by other players or just dissapeared for whatever reason in many of the strategies (but not for a select few)
- It's just variance and the currently short term losing strategies will roar back to life soon enough then we will forget this ever happened
Could be a bit of all 5.
With regards #2 above is there any way we could get an estimate of how many members are running these? Could we do a poll on the forum or something?
Would appreciate some thoughts on the above if anyone wants to join in the discussion thanks in advance
-
@Joseph-Henderson lay 3rd fav Archie score is 0.25 since forward testing in July for example
-
@Joseph-Henderson might be worth running the forward testing dates you have been tracking them through the Archie score
-
I seem to have had a pretty rubbish month again with a version of LTBF for jump racing and Lay 3rd fav. Just wondered how people have got on in November and any strategies that seem to be going well?
-
@Sean-Myler I was losing money big time but I reversed all the members strategies on here and now I am making money. So if there is a members strategy say back, then lay it. And vice versa. Take it in
-
@Andrew-Gray Fair enough!
Does anyone have any experience using any other lay staking plans? There appears to be a lot of different options both online and on BFBM. Would be interested to hear of any experiences (good and bad!).
-
Loss recovery plans work great... until they don't then you lose your whole bank
Anyone trying to tell you they are good doesn't even understand basic maths and should be avoided at all costs!!!
-
@Adam that’s interesting. I tried that staking plan on a few greyhound strategies im testing and I found it was very sensitive to the average/max lay odds. For example I have one where it is profitable but more like 6-7 average and the strategy made a loss when applying the Lay 1-4 staking plan. Might give it a try with low stakes
-
@Joseph-Henderson I tried the same thing on a greyhound strategy back in 2009... the issue is how exponentially your drawdown is ramped up on a losing run.
No harm in paper trading it to test it though. Maybe I wasn't limiting the top end of the stake enough. I'm planning on going back through all my old strategy ideas at some point to see if I can apply any new knowledge to them. There's probably some undiscovered gold in there somewhere!
-
@Adam I had done some back testing on the Lay 1-4 staking plan on a greyhound automated strategy I was working on. Initial thoughts were it seemed to work quite well as the average lay odds were around 4.30 so it always recovered quite quickly. Any thoughts..?
-
@Correa Personally I just like boring old compounding based on risking a percentage of my bank, but parlay can be good for building bank quicker if you have a high strike rate strategy.
It might work quite well for the way you described that you're trying to track your strategies in our other conversation as well, as you could kind of "partition" the parlay by day and work it so that you basically only risk the initial stake from your first bet of the day.
-
@Joseph-Henderson Potentially, if the person using it doesn't turn it off. Would scare me personally!