Trading for a living | Page 35 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Living in China, trading for a living 18 replies

1 trade per pair per year 6 replies

multiple systems per pair/multiple pairs per system? 3 replies

Trading 24 hours per day, 5 days per week 14 replies

Attachments: Trading for a living
Exit Attachments

Trading for a living

  • #681
  • Nov 19, 2025 2:57pm Nov 19, 2025 2:57pm
I asked Perplexity the following:

How can I best plan to cover my monthly living expenses from trading, given that I have no financial reserves to bridge months without profitability, and the general consensus in trading is that a constant minimum monthly profitability cannot be guaranteed?

Answer:

It is extremely risky to plan to cover your monthly living expenses from trading if you have no financial reserves or buffer, because trading income is highly variable and there is no guarantee of meeting a minimum monthly profit. Most experienced traders and financial planners advise against relying solely on trading to fund ongoing living costs without a safety net to absorb inevitable drawdowns and losing streaks.

Key Challenges with No Financial Reserves

- Trading income is erratic, and profitable months can be followed by losses or break-evens; unlike a salary, there is no minimum or safety net.
- Without a financial cushion, even a short series of unprofitable days in a single month puts you at risk of failing to meet basic expenses and could force withdrawal from trading at the worst moment.
- Market conditions, slippage, and unexpected events can make any trading edge unreliable in the short-term, further increasing income unpredictability.

Best Practices If You Must Depend on Trading Income

- **Detailed Budgeting:** Rigorously track all expenses and keep lifestyle minimal; knowing your exact financial needs provides a clear target.
- **Risk Management:** Use strict loss limits (daily, weekly, monthly) and set aside funds instantly from any profits to cover near-term expenses before compounding back into your account.
- **Income Smoothing:** Take profit withdrawals only after exceeding a baseline account value (“high water mark” principle), and refrain from withdrawals in loss months to let your account recover.
- **Contingency Planning:** Seek additional part-time income sources, freelancing, or temporary work to bridge unprofitable periods.
- **Avoid Overleveraging:** Using leverage to “make up” for down months is extremely risky; keep risk per trade low and avoid emotional reactions to performance swings.
- **Professional Support:** Consult with a financial adviser or other experienced traders to add external discipline and advice.

General Consensus and Recommendations

Most traders and financial experts strongly recommend accumulating 6–12 months of living expenses in cash reserves, completely separate from your trading account, before attempting to rely on trading as your sole source of living expenses. This buffer dramatically reduces the psychological pressure that can sabotage trading performance and provides a fallback when a string of losses inevitably occurs.

In summary, while it is technically possible to cover living expenses directly from trading, this is a very high-risk path with significant chances of personal and financial distress. Building a financial buffer and diversifying income streams is highly advisable before attempting to sustain yourself solely on trading.
  • #682
  • Nov 19, 2025 3:36pm Nov 19, 2025 3:36pm
Memo to myself. It seems far easier to sell people the dream of making a living from trading instead of doing it.

I still wonder if and how it's possible despite all these seemingly compelling arguments against it.

It might be easier to do this starting as a millionaire in the first place (to make a basic living from trading) but in my current situation with about 25-50k I could potentially deposit in a personal trading account and a system not proven yet to show consistent results (well 3,5 months) it’s probably not going to happen to make min. 5k/month.
  • #683
  • Edited 2:37am Nov 20, 2025 2:21am | Edited 2:37am
Quoting Noisy
Memo to myself. It seems far easier to sell people the dream of making a living from trading instead of doing it. I still wonder if and how it's possible despite all these seemingly compelling arguments against it. It might be easier to do this starting as a millionaire in the first place (to make a basic living from trading) but in my current situation with about 25-50k I could potentially deposit in a personal trading account and a system not proven yet to show consistent results (well 3,5 months) it’s probably not going to happen to make min....
No one wants to hear the Truth about trading, that's why Guru's keep making money by selling Poor Buggers THE DREAM

No one wants or can Handle the Truth
  • #684
  • Nov 20, 2025 2:35am Nov 20, 2025 2:35am
Quoting T721
{quote} No one wants to hear the Truth about trading, that's why Guru's keep making money by selling Poor Buggers THE DREAM
Attached Image (click to enlarge)
Click to Enlarge Name: Screen Shot 2025-11-20 at 08.35.03.png Size: 91 KB
  • #685
  • Nov 20, 2025 2:42am Nov 20, 2025 2:42am
Quoting Noisy
Memo to myself. It seems far easier to sell people the dream of making a living from trading instead of doing it. I still wonder if and how it's possible despite all these seemingly compelling arguments against it. It might be easier to do this starting as a millionaire in the first place (to make a basic living from trading) but in my current situation with about 25-50k I could potentially deposit in a personal trading account and a system not proven yet to show consistent results (well 3,5 months) it’s probably not going to happen to make min....
Inserted Video
  • #686
  • Nov 20, 2025 2:55am Nov 20, 2025 2:55am
Quoting T721
{quote} No one wants to hear the Truth about trading, that's why Guru's keep making money by selling Poor Buggers THE DREAM
Underlying which there is a deep truth, observed already by the Romans. “Mundus vult decipi, ergo decipiatur” or “The world wants to be deceived, so let it be deceived.”

Even St. Augustine, whose thoughts have had significant impact on European history, noted this tendency in people. Not cynically, as some have used it, but as a deep truth about human beings.

Another useful proverb in this context is "If the only tool you have is a hammer, then every problem is a nail."

In other words, how is our view of the markets and their potential shaped by our presuppositions, beliefs, as well as desires and fears?

If, like myself, one has a strong quantitative background, then one can easily believe that better quantitative modelling should lead to improved profitability. There may be some truth to that, but what is it? My own answer is to trade NQ on the 1 minute chart with barely any indicators at all. "No guts, no glory," or the Russian proverb "Those who don't risk don't drink champagne," are closer to how I relate to the market.

Now, to me all these things make the markets and day trading one of the most uniquely fascinating phenomena one can participate in.
  • #687
  • Nov 20, 2025 3:12am Nov 20, 2025 3:12am
Quoting medici
{quote} Underlying which there is a deep truth, observed already by the Romans. “Mundus vult decipi, ergo decipiatur” or “The world wants to be deceived, so let it be deceived.” Even St. Augustine, whose thoughts have had significant impact on European history, noted this tendency in people. Not cynically, as some have used it, but as a deep truth about human beings. Another useful proverb in this context is "If the only tool you have is a hammer, then every problem is a nail." In other words, how is our view of the markets and their potential shaped...
I like 1m too. You will be quick in profit or stopped out. Feedback loop is quick, possibility quicker correct mistakes and learn to trade.
In London session it makes more sense for me to trade Gold, its 24h market. Simple, supports to resistance, resistance to support trades.

Today so far :
Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 10 KB
I might be wrong , I have been wrong before
  • #688
  • Nov 20, 2025 10:42am Nov 20, 2025 10:42am
To accurately estimate the probability of achieving a certain minimum monthly performance, you need to consider two aspects: the size of your original daily P/L dataset and the number of simulations (iterations) you run.

Original Daily P/L Data (Input)

There is no fixed "minimum" number of daily P/L values, but rather a need for a dataset that is representative of the true underlying process and large enough to capture potential extreme events or rare scenarios.

  1. Rule of Thumb: A common, general statistical guideline suggests a minimum of around 30 data points for certain statistical assumptions (like the Central Limit Theorem) to hold reasonably well.
  2. For Financial Data: Financial data often has heavy tails and can be highly skewed, which means a larger sample is usually required to capture the full range of possible outcomes accurately.
  3. Practical Recommendation:

    1. For a simple, initial estimation, a few months of daily data (e.g., 60-100 data points) might give a rough idea.
    2. For reliable and accurate risk assessment, especially for extreme percentiles (e.g., 95% or 99% probability of loss), you would ideally want a much larger historical dataset, potentially several hundred or even thousands of data points spanning different market conditions (bull, bear, volatile, calm). A small sample size might fail to capture the true structure of the data and lead to an inaccurate distribution.

Number of Monte Carlo/Bootstrapping Simulations (Iterations)
Once you have your original data, you use Monte Carlo or bootstrapping to generate a distribution of possible monthly performance outcomes by resampling or simulating many scenarios. The number of simulations you run determines the precision of your final probability estimate, not the accuracy of the underlying distribution itself (which comes from the input data).

  1. For General Accuracy: Between 500 to 1,000 simulations is often considered a good starting point for stable estimates of mean and standard deviation.
  2. For High Precision: To be highly confident in the extreme percentiles (e.g., a 99% probability of achieving a minimum performance), you might need 10,000 or more simulations.

In summary: Focus on acquiring as much representative historical daily P/L data as possible to build a solid foundation for your analysis, and then run several thousand (e.g., 1,000 to 10,000) simulations to ensure the stability and precision of your results.

  • #689
  • Nov 21, 2025 5:10am Nov 21, 2025 5:10am
Quoting medici
In my experience, Monte Carlo Simulation beats the ratios by far, in particular in situations like this with limited data. You could simply tell Perplexity Research something like this: I have a profitable day trading strategy with three months of daily returns data. How likely is it to be able to deliver 2% return monthly with a certain target maximum drawdown?
I didn't know until now that you could run a Monte Carlo simulation using only historical daily PnL data. I was only familiar with these Monte Carlo simulation tools where you can't use any real existing values at all, but you have to define things yourself like starting capital, win rate, risk per trade, RRR, number of iterations, etc.

I asked Perplexity if there were any other things besides daily PnL that I needed or had to consider.

Answer:

Using only historical daily P/L implicitly assumes:

  1. Daily returns are IID (independent and identically distributed)
  2. Your distribution does not change over time
  3. Autocorrelation is zero (no streakiness, no volatility clustering)
  4. Regime changes don’t matter
  5. You have enough data to represent tail risks

These assumptions are often invalid for real trading strategies.

So you can do Monte Carlo with just P/L values, but your risk estimates may be misleading.

I currently only have 45 days of daily PnL data. I assume that's hardly enough to determine the probability of a certain minimum monthly return at say a 95% confidence level.

Attached File(s)
File Type: csv DailyPnL.csv   2 KB | 58 downloads
  • #691
  • Nov 21, 2025 8:17am Nov 21, 2025 8:17am
Quoting Noisy
{quote} I didn't know until now that you could run a Monte Carlo simulation using only historical daily PnL data. I was only familiar with these Monte Carlo simulation tools where you can't use any real existing values at all, but you have to define things yourself like starting capital, win rate, risk per trade, RRR, number of iterations, etc. I asked Perplexity if there were any other things besides daily PnL that I needed or had to consider. Answer: Using only historical daily P/L implicitly assumes: Daily returns are IID (independent and identically...


Yes, but based on available historical data, Monte Carlo simulation is more robust for forward-looking risk assessment because it explicitly models uncertainty and can incorporate tail risks that the Sharpe ratio ignores. The Sharpe ratio remains valuable as a standardised, easy-to-interpret metric for historical performance comparison, but with its own limitations. Also, Monte Carlo's flexibility allows you to stress-test assumptions and model scenarios beyond observed history.
  • #692
  • Edited 9:34am Nov 21, 2025 9:21am | Edited 9:34am
Quoting medici
{quote} Yes, but based on available historical data, Monte Carlo simulation is more robust for forward-looking risk assessment because it explicitly models uncertainty and can incorporate tail risks that the Sharpe ratio ignores. The Sharpe ratio remains valuable as a standardised, easy-to-interpret metric for historical performance comparison, but with its own limitations. Also, Monte Carlo's flexibility allows you to stress-test assumptions and model scenarios beyond observed history.
Thanks. Very interesting topic. I ran a simulation using AI and my existing data of daily PnL values, and this is what I got.

I will now run the Monte Carlo simulation using this column. I am assuming 21 trading days in a month and performing 100,000 simulations to estimate the probability of achieving a monthly return of +2% or more.
Attached Image (click to enlarge)
Click to Enlarge Name: 01.png Size: 9 KB
Based on the Monte Carlo simulation, here is the estimate for achieving a minimum monthly return of +2%. The simulation assumes a normal distribution of daily returns, using your historical data for the parameters.

Monte Carlo Simulation Results

The estimated probability of achieving a minimum monthly return of +2.0% is: 84.41%

Simulation Details

1. Historical Daily Statistics (from your data):

Daily Mean Return: 0.3935%
Daily Standard Deviation: 1.3187%

2. Simulation Parameters:

  1. Number of Simulations: 100,000
  2. Trading Days per Month: 21

3. Simulated Monthly Return Distribution:

The chart displays the distribution of 100,000 simulated monthly returns. The red dashed line marks your target return of +2.0%, and the shaded area represents the successful outcomes (+2.0%). The simulation suggests that based on your past performance (u = 0.3935%, σ = 1.3187%), your monthly returns are highly likely to exceed the +2.0% threshold.

  • #693
  • Nov 22, 2025 1:26am Nov 22, 2025 1:26am
Quoting Noisy
{quote} Thanks. Very interesting topic. I ran a simulation using AI and my existing data of daily PnL values, and this is what I got. I will now run the Monte Carlo simulation using this column. I am assuming 21 trading days in a month and performing 100,000 simulations to estimate the probability of achieving a monthly return of +2% or more. {image}Based on the Monte Carlo simulation, here is the estimate for achieving a minimum monthly return of +2%. The simulation assumes a normal distribution of daily returns, using your historical data for...
Good stuff. And what is the distribution of the maximum drawdown over the 252 trading days of a year?
  • #694
  • Edited 7:28am Nov 22, 2025 7:05am | Edited 7:28am
Quoting medici
{quote} Good stuff. And what is the distribution of the maximum drawdown over the 252 trading days of a year?
Mean MDD: 8.08% | The average maximum drawdown observed across all 100,000 simulated yearly scenarios.
Median MDD: 7.65% | Half of the simulated years had a maximum drawdown of 7.65% or less.
95th Percentile MDD: 12.64% | In the worst 5% of simulated years, the maximum drawdown did not exceed 12.64%. This is often used as a conservative, worst-case risk estimate.

--
However, I'm not entirely sure how meaningful this information is because:

1.) Very limited number of daily P&L values
2.) Very limited time frame (possibly just a good phase/optimal market conditions)
3.) No consideration of other metrics such as average number of trades per day, win rate, average win, average loss?

Furthermore, the strategy involves gradually increasing the position sizes of trades after previous losses in order to get out of the drawdown somewhat faster (albeit only up to a certain limit - not a classic ruthless Martingale). But it could perhaps play a role that such a Monte Carlo simulation does not take this into account (not sure).
  • #695
  • Nov 22, 2025 8:20am Nov 22, 2025 8:20am
I still don't quite understand how and where Perplexity arrives at such an assessment, and especially these overwhelmingly positive results based on random entry orders. The PDF cites 103 sources, but I've looked at them. Nowhere is a strategy like the one in the Perplexity PDF mentioned. Then it mentions a "comprehensive backtest" it allegedly carried out. While the trailing profit strategy is working now, despite using several different parameters and a tick-by-tick backtest on the Nastaq 1-minute chart for the last year - I'm still nowhere near getting those results!

In most scenarios (backtest runs), it ends with a brutally high drawdown and negative profitability. A few backtests have yielded positive results so far, but these are more like 2 out of 100, so anything but robust and therefore seem to be pure chance rather than a conclusion that random entry orders with certain risk/profit management could sustainably lead to profitability, quite apart from these utopically positive results from the PDF.
  • #696
  • Edited 10:36am Nov 22, 2025 9:08am | Edited 10:36am
Quoting Noisy
{quote} Mean MDD: 8.08% | The average maximum drawdown observed across all 100,000 simulated yearly scenarios. Median MDD: 7.65% | Half of the simulated years had a maximum drawdown of 7.65% or less. 95th Percentile MDD: 12.64% | In the worst 5% of simulated years, the maximum drawdown did not exceed 12.64%. This is often used as a conservative, worst-case risk estimate. -- However, I'm not entirely sure how meaningful this information is because: 1.) Very limited number of daily P&L values 2.) Very limited time frame (possibly just a good phase/optimal...
That sounds good.

Yes, the small set of data reduces reliability, but with that limitation this is as good as it gets in terms of answering your questions about return and drawdown in an evidence-based way.
  • #697
  • Nov 22, 2025 12:34pm Nov 22, 2025 12:34pm
Quoting Noisy
I still don't quite understand how and where Perplexity arrives at such an assessment, and especially these overwhelmingly positive results based on random entry orders. The PDF cites 103 sources, but I've looked at them. Nowhere is a strategy like the one in the Perplexity PDF mentioned. Then it mentions a "comprehensive backtest" it allegedly carried out. While the trailing profit strategy is working now, despite using several different parameters and a tick-by-tick backtest on the Nastaq 1-minute chart for the last year - I'm still nowhere near...
I believe the issue is that Perplexity uses synthetic 1m data rather than historical tick-by-tick data.
  • #698
  • Nov 22, 2025 9:04pm Nov 22, 2025 9:04pm
Quoting Newtrader025
That’s the harsh reality! There is no guarantee for your hard-work and commitment! I don’t know about your learning process! But for me, it was all about simulation for building my edge! It took more than 3.5 years, to understand the logic of multiple time frame based movements! But believe me, after 1 year of back testing, I though; I learned everything; but the reality was diffident! I tried more and more and failed most of the time! Till now the area of my improvement is huge! By the way, from my 1st day of trading till now only focus on only...
I adore this thread concept. Treating trading like a job, not a lottery, with set routines, set hours, and set risk, was what really benefited me. I mostly trade indices and currencies. I used to worry about my inconsistent income, but creating a quarterly budget rather than a monthly one really helped.
https://topasiafx.com
  • #699
  • Nov 23, 2025 5:52am Nov 23, 2025 5:52am
Quoting medici
{quote} That sounds good. Yes, the small set of data reduces reliability, but with that limitation this is as good as it gets in terms of answering your questions about return and drawdown in an evidence-based way.
Yes, I agree. That's definitely a good starting point.

Quoting medici
{quote} I believe the issue is that Perplexity uses synthetic 1m data rather than historical tick-by-tick data.
Yes, I think so too.

--

I feel like I'm trying to master too many different things at once again lately. It's probably just my nature, being a very inquisitive person with many interests. I really need to focus on something in trading that has the potential to help me achieve my goals. The ORB system, while admittedly far from perfect, seems to be working reasonably well so far. I haven't bought the other system for NinjaTrader yet. The annoying thing is that it requires a monthly subscription; I don't think it's even offered as a one-time payment. I already have enough monthly expenses, but of course, if it could be profitable, it might pay for itself quickly. I'm going to try another prop trading challenge with it at the beginning of December. Ultimately, I need a lot of capital to make the necessary money from the targeted 2% per month.
  • #700
  • Nov 23, 2025 8:43am Nov 23, 2025 8:43am
Quoting Noisy
{quote} Yes, I agree. That's definitely a good starting point. {quote} Yes, I think so too. -- I feel like I'm trying to master too many different things at once again lately. It's probably just my nature, being a very inquisitive person with many interests. I really need to focus on something in trading that has the potential to help me achieve my goals. The ORB system, while admittedly far from perfect, seems to be working reasonably well so far. I haven't bought the other system for NinjaTrader yet. The annoying thing is that it requires a monthly...
It's good to see you learning things And while you are perhaps juggling too many things when it comes to learning, you are remarkably one-eyed with you trading ambitions, it seems. Portfolios of algos beat single algos any day for someone with your requirements. And the Swiss provider you are referring to looks like a good start for building a portfolio. But of course you would have to do your own research about this.
2 traders viewing now