Otto does not use grid, martingale or averaging methods, and never adds to a losing position. This page is the reason, in arithmetic.
A martingale doubles its stake after every loss so that the first win recovers everything. Start with one unit. Lose six times and the seventh stake is 64 units, and 127 units are on the table. Lose eight times and the ninth is 256. Every strategy has runs of eight losses; a strategy with a 45 per cent win rate will see one in most years. At that point the martingale needs 511 units of margin to continue, and an account that started with 100 does not have it. The broker closes the positions at the bottom, and the years of small wins that preceded the run are gone in a week.
A grid does the same thing more gently. It adds a position every fixed distance as price moves against it, so that a 500-pip adverse move produces ten positions with an average entry 250 pips away from the current price and a floating loss that grows with the square of the move. It waits for a bounce that will bring the basket to breakeven. In a trend, the bounce does not come, and the grid meets the margin limit at the point of maximum loss.
Both methods produce a rising balance curve and a win rate above 90 per cent for as long as they last, because no losing trade is ever closed. That is why they dominate EA marketing. The equity curve, which counts open losses, shows what the balance hides, and the day the two curves meet is the day the account ends.
Otto's approach is the opposite in every respect. Every trade has a stop from the moment it opens, a losing trade is closed at that stop, position size is fixed by risk before the trade and never increased afterwards, and the portfolio controls reduce size in a drawdown rather than adding to it. Otto has losing months. It records them, at the time, in the account. That is what a record with losing months means: the losses were taken rather than carried.