The Just Calculator for Optimal Money Management

If you haven't already seen my replies at https://www.forexfactory.com/thread/...4#post14762344, after a long time spent in research and simulation, along with many breakthroughs, I believe I have cracked the code for the optimal money management in trading. I think it's time to create a separate thread about it and have made it all into a very accessible and user-friendly tool. Introducing TheJustCalculator.com



Imagine you have a good trading strategy, either in backtesting or live trading, for example, a strategy with a win rate of 50% and an average risk-reward ratio of 2. But if you don't have any idea how to manage your capital or risk, should you risk 0.5%, 1%, or 2% per trade? What about drawdowns? What if you lost 6 times in a row? How do you recover from losses? Or if you won 6 times in a row and made a profit, how much should you withdraw? I've created the perfect trading tool to guide you in these scenarios. The most important factor is having a consistently profitable trading strategy, and the calculator will handle the money management for you.

First, go to https://thejustcalculator.com/, take a quick look at the site description and guide, and then return to the calculator. Begin by clicking on the gear icon on the corner right to set up the calculator.



In this example, I set up a forex configuration where one lot equals 100,000 units, the fee is 3.5 USD per lot, and I included settings for maximum leverage, maximum order, minimum order, and order digit. For the maximum drawdown (Max DD) and the maximum risk, this is where you can customize them; for now, just leave them blank. Then click "Apply Settings."

Now, for the main UI of the calculator, imagine you have a spare $10k and you want to trade with it. Obviously, you won't invest it all in the broker. Based on my research, the best approach at the start is to deposit 20% with the broker as our capital and leave 80% in the bank. This is what you will input into the calculator: $2000 for capital and $8000 for the bank, along with our winning rate and average risk-reward ratio. Now, imagine you've found a theoretical potential trade in EURUSD with an entry price of 1.08105 and your stop loss at 1.07960. Input all that and click "Calculate Size."



Here, the calculator will indicate how much to buy or sell in lots and at what minimum leverage. It will suggest you BUY 0.61 LOT, risking 4.71% of the capital, which is $2000. However, in total value, this represents 0.93%, meaning in the case of a loss, it could potentially be $93.06, not including slippage. The calculator also takes into account the fees, with the total fee amounting to 4.96%. Therefore, from the $93.06, the total fees for both entry and exit would be $4.76.

But we're not finished yet. If you scroll down, you'll see sections for "Outcome" and "Calculate Withdrawal." Here, you need to input either your capital balance after closing the trade or the trade's P&L after closing. Let's assume the trade was successful and you made a 2x profit. Enter $200 in the Profit/Loss field, so the new capital will be $2200. Then click "Calculate Withdrawal."



The calculator will also suggest how much to withdraw after a winning trade, accompanied by some encouraging messages. It suggests withdrawing $89.60 of the profit, which means that from your initial winning trades, you can withdraw and enjoy your earnings on real-life things. Just make sure to record that withdrawal in the Bank section, and the calculator will take it into account. The calculator also summarizes all the trade information and provides a CSV text, which you can copy into your personal trading log for easy logging. Once that is done, you click "next," and the calculator updates the capital and balance to the current values.

That is generally how the calculator is used. You can test various inputs, different winning rates, and risk-reward ratios. You can also view simulations that synchronize with the calculator's inputs, like this one.



The calculator is developed entirely in JavaScript, allowing you to download it, review the code and use it offline use. However, if you're interested in delving deeper into the theory behind it, I recommend reading my paper titled "Beyond the Kelly Criterion." In it, I detail all of my research and the steps that led to my final discovery, as well as how I incorporated all of that into the calculator.

https://zenodo.org/records/10681498

It's been a long journey filled with numerous discoveries and theories, and I'm quite satisfied with the results of my findings. I've done my best to make the calculator as user-friendly as possible for traders, including features like the small window version and local storage to keep inputs stored in the browser. I hope the tool will prove useful and beneficial to many traders, both new and experienced. If you have any questions or feedback, please feel free to ask.

Best regards,

Justus L. Ludovic

Interesting… but if size is constantly adjusted to most recent performance you risk overfitting (fat tails, vol clustering). This seems very theoretical and I suspect in practice it’d fold quite quickly. Have you any real world validation?

Thank you for the feedback. Similar to the Kelly Criterion, the calculator only needs a profitable win rate and risk-reward ratio to suggest a position size. In trading, it is challenging to know these two values exactly, so I believe that the statistical average of the last 100 trades is a good estimate.

I'm not sure what you mean by overfitting. The algorithm isn't a machine learning model nor a linear/logistic regression; it doesn't attempt to predict anything. Instead, it adjusts to your current money management situation, such as the current ratio between the capital at the broker you're trading with and your wealth or bank balance. If, after some trading, the capital grows larger than your bank, it suggests risking less on each trade and withdrawing more after each win to secure those profits. Conversely, if the capital decreases during a drawdown, it recommends risking more on each trade and withdrawing less or nothing to recover quickly from the drawdown. This adjustment in size and withdrawal strategy is deduced by an equation that is mathematically optimized for your current win rate, risk-reward ratio, and bank-to-capital ratio. All of this is based on the Kelly framework but in a more sophisticated and safer manner than Kelly's original formula. So, it's not really a predictive model.

Regarding real-world validation, I'm not yet a profitable trader. As I mentioned in my posts on the other thread, I've done a lot of manual and automated trading, both in demo and live. Unfortunately, the win rate and risk-reward ratio of my last 100+ trades in forex and crypto have a negative expected value, so my calculator won't even suggest a position size. However, if I had a profitable strategy, of course, I would use my calculator. Currently, I'm still testing the calculator with penny risks on Binance and in Forex. Just today, I fixed two minor UI bugs. If I become profitable, and made profit using it, I'll definitely share it here.

I get you, I mean overfitting to your results, but I had misunderstood exactly what you were doing to be fair (you need a summary page!). In which case, nice idea - congrats. It's a tough one to test so the proof is in the doing, but I respect the work and thinking. Good luck

I've been searching for a calculator like this for a long time. I'm excited to try it out. However, it might take me several months to accumulate 300 trades.
I'm planning to use it for trading futures instead of forex, so I'll need to tweak a few things.

Also, I think it'd be easier if it's in an Excel or worksheet format.
Anyways, thanks a bunch for putting in all that hard work for free!

You're welcome! I hope it will be beneficial for you. I have already made an Excel spreadsheet in the other thread, but that was for a simpler formula, specifically the Tax Criterion in the paper. This one is more complex, as it requires several layers of the Bisection method to solve multiple equations and find the position size. Therefore, it needs to be coded into a macro and imported into the spreadsheet. I could use ChatGPT's help to translate the code. If I manage to create it, I will definitely share it here. In the meantime, at the end of the calculator, it says 'Copy Trade' before 'Next Trade'. You can paste that directly into any Excel spreadsheet. This will make using Excel in parallel with the calculator more ergonomic.

The calculator is completely free and open source. However, occasionally, after congratulating you on a winning trade, it may prompt you to tip a portion of your profit, but this happens only when you win, never when you lose. Moreover, if anyone, anywhere on the internet, tips the calculator, it checks the blockchain, and if it detects a donation, it will display 'someone tipped x$' and celebrate with a smiling emoji! So, while it's free, it might become bothersome with its requests for tips after winning trades. Still, this is preferable to cluttering the entire page with ads, which could distract users during their trades. If you prefer not to see these prompts, you can download the page for local use and comment out the tipping part in the JavaScript.

Currently, I'm doing further research on the formula for parallel trading. You mentioned that 300 trades would take several months, maybe if done one at a time. However, if there is an opportunity for multiple parallel trades across different assets, this could increase the number. My question remains the same as with a single trade: what is the best money management strategy for parallel trading? This question applies both to this case and also the case of continuous return trading. The Kelly formula, which is the mean divided by the variance, is already known. I am thinking whether I can adapt my formula to a continuous format and extract a profitable trading strategy from that. I will keep you updated if I find anythings useful.

This is good work, I've always worked out personal equity + broker account size + margin requirement and it's worth the effort. Especially useful using your own account rather than a prop firm.
Your next stage is maybe to code it into MQL4 as a trade manager...

How can the calculator be reset to zero to initiate a new series of trades?

Go to settings (the gear icon at the top right) and clear the ROR Log if you wish to start over. It's a log I created to track the Return on Risk of your trades. Once it logs 100 trades, it will automatically recalculate the average winning rate and the risk-reward ratio after each trade. If you want to restart, simply clear it (or copy it to a spreadsheet if you wish to preserve the history), then click "Apply Settings" to start from zero.



You remind me that I should make a button to reset the calculator on the settings. I will work on it on this week-ends.

Thanks

Hi,
I have read through these threads and the paper and like the work. however i can not access the calculator you have linked. Is this still up and running or is there a way to get access.
Thanks

I'll leave these here to help future visitors and readers of this thread.

Hi,
I’ve fixed the problem, my hosting site did some new update, so I need to manually rebuild the site for their new update. The site is accessible now: https://thejustcalculator.com/. You can download the html page, it work for offline use too. In case the site go down again.
@Erebus, I do check my posts from time to time, but I’m not sure if I was logged in since that last online.

Thanks for the update, Morocco, eh? Could be smoking anything there, probably mess with your memory banks.

You're welcome. I don't smoke, but maybe that daily local skefkef sandwich and mint tea combo zapped my memory. lol

Thanks for the update

It's a pleasure to hear you enjoyed the work! I’ve been diving deeper into new theories with the help of AI tools like ChatGPT, Claude, DeepSeek, and the new Grok. There’s still a long way to go with testing through simulations before I can compile everything into a new paper.

In the Beyond Kelly Criterion paper you read, I focused on betting and discretionary trading, covering everything from optimal sizing to risk management, withdrawals, and drawdowns. It’s particularly useful for those with profitable trading strategies involving leverage, stop losses, take profits, win probabilities, and risk-reward ratios—ideal for high-frequency traders, day traders, and scalpers.

Now, I’m exploring the continuous aspect, specifically for cash investing and unleveraged portfolio management. Using the same process outlined in the paper, I’m now working on optimizing the portfolio Sharpe ratio by introducing a dual-portfolio system: a bank portfolio for buy-and-hold assets and a trading portfolio for active rebalancing. The trading portfolio adjusts based on a generalized Sharpe ratio optimization (factoring in cash and all asset units), while the bank portfolio passively accumulates cash and assets through dynamic taxing/withdrawals during rebalancing, without any selling.

I’m also taking it a step further by developing an internal lending system between the buy-and-hold bank and the trading portfolio. This system allows the trading portfolio to access leverage (borrowing cash) and short selling (borrowing asset units) without interest, as everything operates internally within the same portfolio, independent of broker lending or external margin systems. There’s still much to test, simulate, and optimize—especially given the increased mathematical complexity compared to the first paper. But once it’s ready, I’ll be writing a new paper to share the results.

Hi,
Thanks for your effort. I read the paper and seems very interesting, so I tested the calculator using exactly the same values you used in your example. I was surprised to see different results, your risk was $93.06 but for me (with the same inputs) it was $134.25. Please see the attached screenshot.
Any idea why those results are so different?
Thanks!

Hi,

Sorry for the late reply. Noticing “Buy 0.88 LOT” in the screenshot, it looks like you did some configuration (using the gear icon at the top right). I believe you also specified the Max Drawdown (20%). That’s what makes the difference—whether you specify it or leave it empty by default, the calculator handle it differently.




If you look at the paper, the spreadsheet on page 35 shows that the optimal MaxDD for a 50 % win rate and 1:2 RR is 18%. When the MaxDD field in the config is left empty, the calculator simply works through the formulas to find the optimal max drawdown for your edge (win rate and risk/reward ratio). If you test the same input with the field empty and then with 18 %, you get the same risk: $94.21 (I didnt get $93.06, maybe because I changed the code later after my first posts). With 20 % MaxDD, you get a slightly higher risk: $135.02. When you specify MaxDD, the calculator dont try to find the optimal, it will simply calculate risk to your specific MaxDD.