Martingale Revisited
-
- Posts: 260
- Joined: Wed Nov 03, 2010 12:10 am
- Location: Ballygarvan,Cork Ireland T12D2VR
- Contact:
Ive been looking at Martingale - to improve the return on a strategy which is all theory at the mo. I think Martingale has possibilities. If you have a half decent Win/Loss ratio say 2;1 and a return of about 66% of stake on average then Martingale becomes viable does it not? Of course you would need a sequence limit in place. I know many are dead against martingale but I can see an option for it in some circumstances. Next question if its viable how do you implement in excel? What are the difficulties there?
-
- Posts: 3140
- Joined: Sun Jan 31, 2010 8:06 pm
If you can accurately predict the probability why not simply use the Kelly criterion for your stakes.
i have to agree with SB. Martingale is just not worth the agro. I'm not surprised tho' that it still has some traction. A lot of the snake oil salesmen punt Martingale as the *engine* behind their strategy (*a winner is always around the corner* etc).
If you really do have a decent strike rate with an ROI of 66%, then explore Kelly Criterion as this takes into account things such as probability and adjusts to suit conditions (of the bet).
https://en.wikipedia.org/wiki/Kelly_criterion
paraphrasing t0ne75, go for a skydive, absail, freebase... literally look at much safer options!
If you really do have a decent strike rate with an ROI of 66%, then explore Kelly Criterion as this takes into account things such as probability and adjusts to suit conditions (of the bet).
https://en.wikipedia.org/wiki/Kelly_criterion
paraphrasing t0ne75, go for a skydive, absail, freebase... literally look at much safer options!
-
- Posts: 260
- Joined: Wed Nov 03, 2010 12:10 am
- Location: Ballygarvan,Cork Ireland T12D2VR
- Contact:
Its all about the Bank i think.If you apply a sequence Loss limit then going bust might not arise. A massive bank (relative to your stake) might help. I think kelly afaik is not as lucrative (and Im a greedy b...d). To be perfectly honest I have a winning strategy in mind which itself can wipe your bank if you are unlucky to start at a bad time but if you get over the hump of about 3 weeks in then it should win ( A paupers last words!). I was able to backtest the strategy over 1 years data (with a bit of difficulty with constructing the formulas) and the multiple of increase was approx 3.5.
I limited the stake to a maximum 10 fold initial stake increase in a loss sequence ( or days loss to date if lesser and writing off any cumulative losses of previous day). I am sure others have done similar and would like their thoughts before I read that book or go for a pint.
Id love to have the whole thing automated in excel though. Dont fancy 1 to 9pm in front of PC.
I limited the stake to a maximum 10 fold initial stake increase in a loss sequence ( or days loss to date if lesser and writing off any cumulative losses of previous day). I am sure others have done similar and would like their thoughts before I read that book or go for a pint.
Id love to have the whole thing automated in excel though. Dont fancy 1 to 9pm in front of PC.
-
- Posts: 3140
- Joined: Sun Jan 31, 2010 8:06 pm
As long as you have distinct rules (no late flip flopping favourites etc ) and are able to accurately obtain results, there's no reason you shouldn't be able to automate things. Just give it a go that's the only way you'll be able to tell if you've stumbled upon something others haven't.
- MemphisFlash
- Posts: 2222
- Joined: Fri May 16, 2014 10:12 pm
- Location: Leicester
Martingale, fibonnacci, Kelly criterion all strategies that will send you to the poor house.
Yes, but you'll lose a large chunk of your bank when you reach your limit!
You can't take money that's not on the table. The Kelly criterium calculates the optimal stake, why would you want to stake more? Why don't you pressure your tyres to 50 psi? Surely that must be better and safer than 35?
A winning strategy that can wipe out your bank, of course, that's the first thing we all think of when we have a winning strategy, but if you use Kelly it can't wipe out your bank because your stakes reduce proportionally.
It should win anyway, even more by using the Kelly criterium.
What exactly's wrong with Kelly?MemphisFlash wrote: ↑Sun Jun 21, 2020 4:03 pmMartingale, fibonnacci, Kelly criterion all strategies that will send you to the poor house.
Have you considered this Maria esque staking plan that only risks a % of your bank. Potentially you could link up your selections to it and have the stake calculated to return to your v55 status clearing sheet.
You do not have the required permissions to view the files attached to this post.
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
If you want to explore this just do it in excel. You'll be suprised at the length of losing runs even with a strike rate of evens. Was only a couple of years ago that we were all taking about the 23rd (I think) straight losing favourite.
If you do the maths you'll see that you either have a winning strategy or not, staking, like stop losses, are red herrings. Kelly is the only one with credability but needs a great ratings system, and nobody here has one.... or they wouldn't be here.
And let's face it, millions of people have been gambling for centuries, incl some of the brightest minds in maths. If you or anyone thinks they've discovered anything new then they need to do their homework. All they've done is reinvent a broken wheel.
If you do the maths you'll see that you either have a winning strategy or not, staking, like stop losses, are red herrings. Kelly is the only one with credability but needs a great ratings system, and nobody here has one.... or they wouldn't be here.
And let's face it, millions of people have been gambling for centuries, incl some of the brightest minds in maths. If you or anyone thinks they've discovered anything new then they need to do their homework. All they've done is reinvent a broken wheel.