That's ok, as long as the tax is paid. If taxes are due but not paid and subsequently lost in drawdown, with no carry back available, that's going to be a big problem In those jurisdictions you absolutely have to remove the money for taxes at the end of the tax year. Then immediately adjust your position-size/risk-level going forward based on the lower amount in your account after the taxes were removed.
This reminds me of this YouTube i watched recently. In this scenario, the influencer tested a 1:1.3 ( going for a 30% gain per attempt) risk to reward ratio with a 60% win rate risking 23% per trade. Then he proposed to start with $20 and ride it up to $50k+ in about 30 tries ("steps". assuming 30 more wins than losses). Then he set up some algorithm to test this in some automated fashion (a Monte Carlo), for 1000 trials. What happens is this will fail, and lose $20, five times out of 1000. But it will turn $50k+ the other 995 trials. But like others here have said, these stats really have to be met and maintained. When the influencer reduced the win rate to 50%, all else being the same, only 30% (333) of the trials succeeded in making $50k+. All others blew up. I would think you could risk 50% per trade given the OP's expected statistics (win rate + risk ratio) and succeed with a pretty small chance of blowing up a handful of accounts out of a thousand tries. Especially if you get a little lucky and start out strong as the OP has. (66) Testing $20 To $52,400 Strategy 1000 TIMES - Fastest Way To Grow Small Trading Account - YouTube
You got it backward. As the 1%, you need more of us 99% to play, otherwise where and who are you going to take money from?
I think there is a flaw in all these models. They apply probability assuming normal distribution. Why is that we don't question if normal distribution is appropriate for trading data sets? Questioning this underlying assumption is the paradigm shift we need. For example, a normal distribution assumes thin tails. Does this hold for trading? Another one, when in profit our brokers don't call us. But when in loss, liquidation is triggered. Is normal distribution still valid in this case? Again, as humans we endure loss hoping for profit. But we don't want to leave a small profit on the table, as if pick pockets are lurking around. Does this behavior justify normal distribution? I think all these probability tools like Monte Carlo look nice on paper.They may not reflect probable outcomes as we seem to assume. Just my thoughts.
You don't take that seriously, do you? Give me 1000 rolls of the dice craping out only 5 times ...... and I'll OWN Vegas baby. Monte Carlo. Macao, etc.
I'm afraid some fools here actually believed your sarcasm. Sure, let's do exactly what you suggested and completely clean house (eg. trade 1000 lots) by next week.
If you can pull off five winers in a row and double your money every time you can go from 15->30->60->120->240->480->960 contracts. Your broker might impose a position limit on you before 1000 lots, even if you have $1M in margin. You are a discretionary ES trader, i'm guessing you can easily pull off 5 winners in a row? Im an automated trader and my algo can't do it. Except very very rarely.
You could design a Monte Carlo with some pretty realistic parameters reflecting things like average win, average loss, max wins in a row, max losses in a row, max win, max loss, profit factor, etc. You could also take a number of actual trades and reshuffle them a thousand times to see what would happen in a worst case scenario. So yah, i do take stats like this seriously. You start with however much you could stand to lose five times out of a thousand, for example. With really high volatility in outcomes, you would start with a smaller amount at risk for loss. Keep in mind that a risk of 50% does not mean you have two chances. You are risking 50% of a reduced amount, you know, like the half-life of carbon dating. There could be many tries even at a 50% risk per try.