It all comes down to a 0% risk of ruin. Using risk tolerance is an indication of uncertainty, which means you have some work to do. It is possible to raise your conviction level to certain. You will reach your personal extraction potential. Then just have a replenisher bank account with 5x your balance for rare anomalies. Fixed means the Lot size is fixed.
String of losses are not random. Albeit, Trading involves skill*. Without skill, then of course you will deal with a "string of losses". First get skilled. Once you have skill, solve the actual problem, by finding the precursors to a "string of losses" and minimizing it into irrelevance. The reason I say this is the fact the solution mentioned can be described as, two extremes, a neutral point, and increments, a simple scale of first order trade tactics. This type of "quiver", suggests it is "played out", wrt to first order issues. Basic Model theory. Hence why you are asking for alternatives. Perhaps something non parametric in nature or some exotic scaling as an alternative. NB the fact it is a "string" does not mean it needs to dealt with as a "string", with solutions that work on "string" type problems, or solutions that have string like properties, like scaling. Hence why the solution you threw out, and I mean threw out, is so meh. Instead these issues can be handled in the Meta System, or Testing Harness. These second order systems, can have non parametric measurements and controls that are better suited to non random first order events, string like or otherwise. They can act as a bridge between the parametric first order events and the non parametric measurements that inform the control of the trading (first order actions). This is where meta skill is encoded. Does not matter if it is the system or the trader. They both need a Testing Harness or Meta system. Otherwise you are stuck with first order solutions or a mishmash of first and second order solutions in a single model. The latter is not a good foundation for orthogonality, scaling, testing or maintenance. The Test Harness needs to be actively influencing trading in real-time and remain dynamically adaptive to the first order events (trades etc.) as they happen. It is pretty straight forward once you setup the models, and can be setup to solve first order issues with non parametric solutions. It is also how you would evaluate any would be, first or second order solution, parametric or non parametric. Best of luck *Skill means you can deftly APPLY all tools in their variants, at the right time, under all the conditions, dynamically, ignorance not excused.
I don't know how to quantify the % of risk of ruin when the exposure is constantly changing due to the ORM (Ongoing Risk Management) +TLSR (Total lot size reduction). Using these techniques + our occasional manual intervention the TLS is kept small for the account size. Therefore, I consider the possibility of ruin a near impossible event. Of Course there will always be a level of uncertainty in our strategies because we trade many currency pairs but I don't consider it high enough to be concerned about, because all accounts are well funded. Ok I have an idea of what you are doing, personally I don't feel comfortable with trade up sizing. I've considered it and I know it can work but a string of losing trades would stress me out and I don't need the stress.
There is a point at which risk of ruin goes from 0 to 100. Find it and work back from there. Once you've tested 5000+ samples your conviction raises to absolute certainty, this is where the stress fades. Once you have this optimized extractor in production, you are free to continue gradual development that lowers the 0 to 100 threshold for larger extractions. You don't have to trade this way but having the option to is proof of mastery. We think like a company, having every possible way something can be done. Indefinite research and development.
I encourage my students to apply endless development to life in general. Don't you get bored otherwise, doesn't the apathy get to you? Settling is easy, as a society we admire difficult feats. We are programmed to consider the easy route pathetic. Live effortfully.
Doesn't work that way. Every possible way is like saying to count grains of sand. Google's AlphaZero cannot do that either, see Sedol match. Your every possible option is akin to the infinite money theorem. Guess what, it doesn't work. 5000+ is arbitrary. Your posts are equivalent to armchair theory. Show some numbers of statistical significance. Until then, it is relegated to amateur philosophers pondering about world peace and the universe.
I also have many students. Actually they call me ancient grandmaster sensei teacher. I encourage them to ponder the subconscious because within it can the universe be perceived and therefore all truth is already known.
Every possible way is actually a handful of things, it is finite. Those that tell themselves it is infinite use it as an excuse to be lazy. I already did this extensive work. My payment methods post lists 11 target systems. It encompasses the entire landscape. What do you guys think, does 11 sound like inifinity. Stockpredictor, I understand the source of your misconceptions. Don't give into it, you're better than that. Don't be pathetic.
Very well, this is one of the few times I apologise. Forgive me. You're one of the people treading the path in search for the invincible technique. I too am walking in search. So then, infinite is a misnomer. However, from a practical standpoint, it cannot all be encompassed due to computing power, energy limits, etc. For example, a Dyson Sphere may be required in order to generate sufficient energy to calculate effects of not only individual datasets on systems, but combinations of such datasets (which can cause discrepancies), on one of the eleven systems (under the assumption that there are 11). This can be furthered due to overlap between systems, but also due to large numbers of subsystems. What I did as far as practical requirements go, is to look at previous successes in history that appeared, at least in some segments of time, unbeatable. What I mean by that is 'string of wins'. Jesse Livermore along with Munehisa Honma and others, achieved this. I looked for commonality among the remnants of what systems they left behind, and attempted to take portions of those systems to correlate them regardless of condition. This is because their theories allowed for changes in market condition and had such 'string of wins' even then. Yin-yang, bull-bear, up-down, all are words to describe the same fundamental truth. This particular field seemed apparent to me, and I assumed that the fruits from this field would be more easily plucked.