Hi, I have been quietly working on a big performance update, and it’s now live. If you’ve used the strategy generator before, you’ll notice the difference immediately. Here’s what changed and why it matters 🙂

Charts load instantly, no matter the data size

Previously, loading 50,000 bars of intraday data could freeze your browser for several seconds. Now it renders immediately and stays responsive. Zooming, panning, and switching between charts is fluid.

You can now see your full price history

Before, charts were silently cutting off data older than roughly 5–6 years even if history went back to the 1990s. That’s now fixed. Load 30 years of daily data and the chart shows all of it.

Strategy generation is dramatically faster

This is the big one. The generator now runs on all your CPU cores at the same time, instead of just one.

Before 1 core was used, now it can use up to 8! In practical terms: what used to take 60 seconds now takes around 8. And while it’s running, you can still interact with the page normally: filter results, view charts, change settings, without any lag!

What You Need To Do

Nothing. Just reload the page. All improvements are automatic.

If you’re using a multi-core processor (which covers virtually every desktop and laptop made in the last decade), you’ll immediately benefit from the full speedup.

Now I can focus on data: add more pre-loaded instruments and intraday timeframes.

Happy backtesting!

Test it by yourself: https://www.prorealcode.com/prorealquant/

prorealcode-prorealquant-2026-02-18-11_40_17.jpg

Hi,

Thanks! Glad the speedup is noticeable.

That’s a great suggestion and you’re right, running Monte Carlo one by one on each strategy is tedious, especially when you have dozens of candidates to evaluate.

Batch Monte Carlo is definitely something I want to add. The idea would be to let you select multiple strategies (or all filtered results) and run Monte Carlo on the whole batch automatically, then display the grade directly in the results table as an extra column. That way you could sort and filter by Monte Carlo grade just like any other metric, which would make the whole workflow much smoother.

The good news is that with the new multi-core engine, this is now technically feasible without freezing your browser. Before this update, running even one Monte Carlo (100 simulations) was already heavy on a single core, so batching would have been painful. Now that the workload is distributed across all available cores, batch processing becomes realistic.

I’ll add it to the priority list. It fits naturally as the next usability improvement after the performance overhaul.

Thanks for the suggestion!

Hi everyone,

Another update just went live on ProRealQuant. This one addresses the most requested feature from the last changelog: batch Monte Carlo testing. But instead of just batching it, I went a step further.

Auto Monte Carlo, built into the backtest

You can now enable automatic Monte Carlo analysis directly during strategy generation. The option is in Advanced Configuration, General tab, check the “Auto Monte Carlo” box and you’re set.

When enabled, each strategy that passes your filters will automatically go through a Monte Carlo stress test right after its backtest. The MC grade then appears directly in the results table as a new column, no need to manually select strategies one by one anymore.

It works in both single batch mode (Generate) and in Continuous mode. So you can launch a continuous session, go grab a coffee, and come back to a table full of strategies already graded by Monte Carlo.

One thing to keep in mind: since Monte Carlo uses bootstrap randomization (100 synthetic price series generated randomly each time), if you run a second MC analysis on the same strategy, the score might differ slightly. That’s completely normal and expected. A robust strategy will stay in the same grade range across multiple runs.

Improved Monte Carlo scoring

I also reworked the scoring formula. The previous version was too harsh on drawdown.

Here’s what was wrong: the old formula penalized drawdown via two metrics (95th percentile max DD and average DD) with a combined weight of 35%. The penalty curves were set so that 50% drawdown scored zero on one metric and 33% drawdown scored zero on the other. The problem is that this was too aggressive for strategies using reasonable stop losses. For example, a strategy with a 5% stop loss that simply hits 5 consecutive losing trades would produce roughly 23% drawdown by pure math (compound: 1 minus 0.95 to the power of 5 = 22.6%). Under the old scoring, that scenario alone would already score around 50 on the drawdown component, even though there’s nothing wrong with the strategy, it’s just experiencing a normal losing streak within its risk parameters.

The fix: the drawdown penalty is now normalized against an “expected baseline” derived from the average stop loss size observed in the simulations. In other words, the scoring now distinguishes between drawdown that’s a natural consequence of the strategy’s risk profile and drawdown that signals a genuine problem. On top of that, the total weight of drawdown in the overall score has been reduced from 35% to 25%, redistributing weight to the other components.

In practice, this means strategies with tight risk management will no longer be unfairly penalized for normal losing streaks. Strategies that produce abnormal drawdown relative to their stop loss size will still be flagged appropriately.

What you need to do

Nothing, just reload the page. Enable Auto Monte Carlo in the General tab if you want it, and let the engine do the work.

Happy backtesting!

monte-carlo-analysis.png

Hi everyone,

Fresh update today with a nice batch of intraday data and several quality-of-life fixes.

New intraday instruments

This is the big addition. You can now generate and backtest strategies on:

EUR/USD 1H, with data going back to 2010 XAU/USD (Gold) 1H, with data going back to 2010 DAX 1H and 30min, with data going back to 2012 US500 1H and 30min, with data going back to 2012

That’s a lot of bars to work with, which means much better statistical significance for your strategies compared to daily timeframes. More trades, more robust Monte Carlo validation, and better Out-of-Sample testing. This was by far the most requested addition, so I hope you’ll enjoy exploring what these timeframes produce.

Fixes and improvements

Charts now render correctly for instruments with more than 2 decimal places. Previously, Forex pairs and some other instruments with small tick sizes were not displaying properly. This is now fixed.

The timescale on charts has been adjusted for timeframes below 1 day. Intraday charts now display time labels that actually make sense instead of showing daily dates.

Trade duration in the orders list is now expressed in bars rather than “days”. This was misleading on intraday timeframes where a trade lasting 5 bars on H1 is obviously not 5 days. Now the unit is consistent regardless of the timeframe you’re working on.

Default order size for Forex pairs is now automatically set to 10,000 contracts. This reflects a more realistic lot size for Forex and avoids having to manually adjust the sizing every time you switch to a currency pair.

The drawdown scale on equity charts has been refined. Values are now always displayed as positive numbers, which makes the chart much easier to read at a glance.

Thank you all

I want to take a moment to thank everyone who has been testing, reporting bugs, suggesting features and pushing the tool forward. Every piece of feedback counts and directly shapes what gets built next. Since launch, millions of strategies have been generated by the community every week, which is honestly amazing to see. The tool is growing fast and that’s thanks to you.

Now I’d love to see more of you sharing your findings here on the forum too! Whether it’s a strategy with a solid Monte Carlo grade, a particular configuration that works well on a given instrument, or just tips and observations from your testing sessions, this is the place to discuss it. The more we share, the more we all learn.

What’s next

More intraday data for other instruments is coming. If there are specific instruments or timeframes you’d like prioritized, drop a message below. Your feedback directly shapes what gets added next.

Happy backtesting!