[Probability of Bankruptcy] Is your trading rule safe?
In the previous article, I explained the mindset needed to stay profitable in trading.
【Trading】What are win rate, RR, and PF?
However, simply having a profit factor above 1 is not enough as a trading rule for winning.
You need to consider the allowable loss per trade.
Yobinori's videos are very helpful.
100 times more likely to go bust【Bankruptcy issue】
Everyone can understand! VarlSAR bankruptcy probability
The bankruptcy probability table for VarlSar is famous, but if the loss tolerance is not written, the meaning changes.
In 1992, Norris Varsala published “Money Management Strategies for Futures Traders,” which includes a table showing the bankruptcy probability when starting with $1 and a loss of $1, varying risk-reward and win rate. There are also tables for starting with $2, $3, $4, $5, and $10.
If you search for Money Management Strategies for Futures Traders, you can view pages with the original content, so please refer to them.
Excerpts from the table including realistic win rates and risk-reward ranges
Also, the bankruptcy probability formula exists only when risk-reward is 1; in other cases, it is calculated via simulations.
Online articles describe it with probabilistic recurrence relations, making it seem as if a formula exists (in most cases, it is explained using characteristic equations, which actually apply to adjacent terms).
It feels like, having studied statistics, that the price staying within ±2σ of Bollinger Bands is within a 95% probability.
【Bollinger Bands ①】About standard deviation
There is a calculation to determine bankruptcy probability only when risk-reward equals 1.
Quoted from Yobinori's video
Also, the common point when risk-reward is not 1 is that
Suppose there is a method with a bankruptcy probability of 1% for a certain amount of capital.
If you halve that capital (double the loss tolerance), the bankruptcy probability jumps to 10%.
If you want a larger return and take on too much leverage, you are mathematically prone to bankruptcy.
Generally, in a method with bankruptcy probability R, increasing capital by a factor S makes the bankruptcy probability R^S.
To explain why intuitively, for example, if you trade with 1,000,000 yen of capital using a method with 10% bankruptcy probability, you would go bankrupt and then continue with another 1,000,000 yen the same way, or you could start with 2,000,000 yen and go bankrupt—the outcome is not different.
・2,000,000 yen → 1,000,000 yen → 0 yen, the same as above, so 1%
However, practically, if you lose half of your investment funds, to recover you would need to double your stake.
If your funds shrink to 25%, you would have to quadruple them to recover.
If funds shrink significantly, recovery becomes nearly impossible.
In that sense, the bankruptcy probabilities discussed so far are the probability of capital reaching zero, so they should be viewed more stringently. You need ways to limit drawdowns.
Warren Buffett's
Rule 1: Do not lose
Rule 2: Do not forget Rule 1
seem to express the importance of bankroll management.
【Capital Management】The concept of units
【Markets】Where should you place your stop loss?
In summary, no matter how high your win rate or PF is, if you increase the per-trade loss tolerance too much, you will be bankrupt.
Even if you have a profitable trading system, going bankrupt midway defeats the purpose.
Conversely, if you set loss tolerance too small, you won’t make money.
You should create a trading rule that balances risk and ensures you do not go bankrupt while taking on some risk.
To illustrate with shogi, no matter how many checks you give, if your king piece is captured, you lose.
If you mismanage defense, you can be checkmated quickly.
Past verification data for trading rules using ATR is now being published!
