5d • Strategy Updates
Parallax Update
We're releasing Parallax v2.5 today, and I want to be direct with you about what this release is and why it took as long as it did.
The Parallax you've had for the past several months stopped performing reliably. We saw it, you saw it, and rather than apply surface-level patches, we took the strategy back to the drawing board. We spent weeks analyzing its behavior across a three-year backtest to understand what was actually failing — not just where the P&L went sideways, but why. Version 2.5 is the result of that work.
We identified three issues, each compounding the others:
1) The trading session was too narrow. The earlier version cut off the first hour of the U.S. open to avoid morning volatility. Our analysis showed that assumption was wrong — the 9:35–10:30 ET window is actually one of Parallax's most productive hours. The strategy's regime-detection layers handle morning conditions well. The new default session is 9:35 AM – 3:30 PM ET, which restores the productive early-session hours.
2) The stop-loss multiplier was wider than the signal accuracy required. With a 60%+ hit rate, Parallax doesn't need an aggressive stop distance to absorb adverse moves. We tightened it modestly (from 2.0 to 1.75 ATR), which reduces average loss size without significantly affecting profit-target hits.
3) Certain time-of-day windows were systematically less productive than the rest of the session. We added a Clock-Side Filter that suppresses entries during those specific hour/direction combinations. The filter is asymmetric — at each blocked hour, only one direction is suppressed; the other continues to trade. It's enabled by default and fully reversible via a single toggle.
The 3-year backtest at the v2.5 configuration produces $53,395 net P&L, profit factor 1.34, win rate 62.5%, and a maximum drawdown of -$7,550. Those numbers describe the rebuilt strategy's behavior across a span that included both choppy and trending market regimes. They are not a forward-performance guarantee.
A few things I want to make clear:
Parallax is one strategy in a diversified suite, and that matters more than any single improvement we ship to any single strategy. No individual strategy is going to perform well in every regime. Parallax is built around mean-reversion and reversion-of-extension setups, which means there will be regimes where it's the right tool and regimes where it's not. The suite is designed so that when one strategy is mismatched to the current regime, others are positioned to carry the load. We've seen this play out in live trading already — there have been weeks where Parallax struggled while Volturon and Quantivus thrived, and vice versa. That's the diversification working as intended.
Which is the long way of saying: please don't size into Parallax v2.5 expecting that the backtest numbers I quoted above will be your experience in every regime. Your forward experience will average something similar over time, but month-to-month and quarter-to-quarter it will vary. The suite is meant to smooth that variation; no single strategy can.
Go to https://volturon.com/software-downloads to download the revised software and user guide. After import, verify the four key settings: StopATRMultiplier = 1.75, SessionStartTime = 09:35, SessionEndTime = 15:30, and Enable Clock-Side Filter = On.
If you run into any issues installing or have questions about the new behavior, reach out directly and we'll work through it with you.
Thank you for your patience during this process.
PS: Nexum and Nodalis have also received updates, so be sure to download and import them as well.
2
0 comments
Parallax Update
powered by
skool.com/futures-trading-group-7221
Welcome to the Futures Trading Group, an exclusive community dedicated to the mastery and advancement of AI-driven automated futures trading.
Suggested communities
Powered by