Goldman Sachs’ mid-day wrap offers a mix of insights into the market’s performance, with some positive and negative trends emerging. Despite the S&P 500’s attempt to notch its first three-day win streak in a month, extreme rotational elements are at play, with Goldman’s Momentum Pair getting torched and Software experiencing its second-largest day of outperformance versus semis in the last year.

The High Beta Mo basket is down by 9% after finishing the first half of 2026 with a record +57% run, as both winners and losers are moving in the wrong direction. Goldman’s team flagged that seasonality is poor for momentum, with July underperformance creating prior summer drawdowns, often driven by rallies in the short leg rather than weakness in the longs. GSPRHIMO has finished lower in every July since 2021.

Activity levels are indicating a holiday week, with the market tracking for its slowest session since Memorial Day. However, panic levels on the desk remain subdued despite sharp moves under the hood.

Goldman’s Franchise segment sees overall activity levels up by 23% compared to the trailing two weeks, running ahead of broader market volumes that are down by 16% versus the 10-day moving average. Both high-frequency traders (HFs) and long-only (LOs) are running in the same direction, with HFs up by 6% and LOs up by 3%. Demand for Communication Services, Industrials, and Consumer Discretionary is offsetting supply in Technology, while demand across Energy and Utilities is outweighing supply in Technology.

Overall, the market’s performance suggests that rotational shifts and seasonality concerns are dominating the landscape, with Goldman’s Momentum Pair and Software sector experiencing significant outperformance versus semis. Meanwhile, High Beta Mo and Franchise activity levels indicate a mixed picture, with some sectors seeing increased demand while others struggle to keep up. As always, investors are advised to stay vigilant and adapt their strategies accordingly in this dynamic market environment.

Leave a comment