/ Market Report / By

R-50 Market Report

November 2024

Market Report

After achieving above-average returns in the summer, the majority of our products entered a phase of consolidation during October with a below-average return of ±2%. Markets shifted in anticipation of the US election, projecting a Trump victory as likely, leading to a significant shift into the USD. The so-called “Trump Trade” emerged, resulting in a firm trend across USD-related FX pairs leading up to the election. By then, EURUSD had already declined by 4% over a one-month period, a significant drop compared to the pair’s usual dynamics. Similar patterns appeared in other pairs across the FX segment, leaving non-USD- correlated pairs with minimal volatility.

With the election approaching and the Fed Interest Rate announcement on the calendar, this movement intensified, erasing 4% of EURUSD within just seven days. This so-called correlation breakdown is a rare phenomenon, usually associated with major market shifts and/or black swan events. During such events, the usual correlations collapse into one focal point, in this case, the dollar, making diversification within FX more challenging.

Despite being constructed from mostly uncorrelated strategies (beta < 0.5), four of our universe of 100+ active subsystems were caught up in this, opening exposure on the same side of the USD (short) across multiple pairs. To reduce exposure and mitigate risk, the affected subsystems were initially hedged and later took controlled losses when a stronger move became evident. These subsystems, despite their long and solid uncorrelated track records, experienced a correlation collapse. This is visible in the chart below, which shows the past two years of live performance for these subsystems:

The affected subsystems have been deactivated and will be replaced once volatility returns to a medium- term average. Looking ahead, exposure risk is now significantly reduced, minimizing the potential for further losses.

R-50

This was the first time R-50 had hit a stop-loss on multiple subsystems in their 2 to 2.5 years of live combined track records. This marked a decline of 4–6% in net performance from October highs, depending on the product, erasing approximately 1.5 months of profits. The negative performance extended across several of our strategies.

Our primary goal at Ridge Capital Solutions is to provide long-term, resilient alpha strategies with limited downside risk. This event demonstrated how effectively our risk management framework handles extraordinary market phases. Although this represents a small setback in terms of overall performance, it underscores our robust risk management methodology, which focuses on controlling exposure and limiting risk.

In reality, sustainable trading requires acknowledging realized losses. We are collecting extensive data on every market movement and corresponding portfolio reactions to continuously enhance our products through real market feedback. A roadmap of near-future improvements is outlined below.

Roadmap

Our roadmap for the coming weeks includes the following. Note that these improvements will be implemented across all existing systems:

  • R-50 Dynamic correlation/covariance adjustment across subcomponents (single assets traded on subsystems) at 15-minute intervals.
  • Implementation of additional trading logic for market regime detection and active trade filtering using
  • Hidden Markov Models (HMM) to predict and identify market regime shifts that may lead to correlation breakdowns.
  • Introduction of non-mean-reverting strategy pools to replace mean-reverting strategies during trending market phases.
  • Reworking subcomponent risk management to prevent clustering of stop-loss events while maintaining tight risk constraints.
  • Potential enhancement and introduction of an automated hedging module.
  • Significant expansion of engineered features in our machine learning model training files.
  • Additional improvements to follow.