A Congressman Bought Defense, Banks, REITs, and Tech in the Same Week. The Real Signal Was Simpler Than You Think.

The Setup

On May 18, 2026, Rep. William R. Keating filed his periodic transaction report under the STOCK Act, disclosing four purchases made on May 7 ¹. Each position was valued between $1,001 and $15,000 and held in his IRA. The buys spanned JPMorgan Chase (JPM), Northrop Grumman (NOC), Simon Property Group (SPG), and VMware LLC (VMW). He also purchased positions in Otis Worldwide, T-Mobile, and Treasury bills on the same day ².

Keating sits on the House Armed Services Committee and the House Foreign Affairs Committee ³. That committee assignment makes the NOC purchase an easy headline, and platforms like Capitol Trades flagged it immediately. But before jumping to conclusions about insider knowledge, consider the full picture: he also bought the largest U.S. bank by assets, the largest retail REIT in the country, and corporate notes from a Broadcom subsidiary. This is not a concentrated bet on defense intelligence. It looks like a diversified quality allocation.

The more useful question is not whether Keating has an information edge. Research from the National Bureau of Economic Research and George Washington University suggests that aggregate congressional trading performance has not consistently beaten the market since the STOCK Act passed in 2012 ⁴. The useful question is: do these four companies share a financial profile that is systematically testable?

How to Think About This

Start with what these companies actually do. JPMorgan Chase earns $306 per share and holds a consensus Buy rating, with Q1 2026 EPS of $5.94 beating expectations ⁵. Northrop Grumman builds B-21 bombers and Sentinel ICBMs, carries a backlog exceeding $95 billion, and operates in a defense budget environment where spending proposals have reached $1.5 trillion for fiscal 2027 ⁶. Simon Property Group runs malls and premium outlets at 96% occupancy with NOI growing 6.7% year-over-year, and recently raised its dividend 7.1% ⁷. VMware LLC now operates as a Broadcom subsidiary; Keating purchased its corporate notes (5.572% coupon, due 2036), not equity ².

Look at what connects them. Each company generates reliable cash flow in an industry with structural demand. Banks intermediate credit regardless of which party controls Congress. Defense contractors fulfill multi-decade government contracts. Retail REITs collect rent from tenants locked into long-term leases. Enterprise software infrastructure runs whether or not AI hype persists. The through-line is not sector exposure; it is durable profitability and cash generation.

In the strategy builder, we can test this thesis by screening for the financial characteristics these picks share. We set return on equity to at least 15% to capture companies that generate strong profits relative to shareholder capital. We require an operating margin of at least 20% to filter for businesses with pricing power and cost discipline. We cap the debt-to-equity ratio at 2.0 to avoid companies that have leveraged themselves into fragility. And we require free cash flow per share of at least $3 to ensure these companies actually convert profits into cash, not just accounting earnings.

The control strategy uses the same universe and filter thresholds but ranks stocks primarily by 6-month price momentum (80% momentum, 20% fundamental). The thesis strategy flips that ratio (40% momentum, 60% fundamental), giving heavier weight to the quality characteristics rather than recent price action. This comparison isolates whether the "quality-first" approach that Keating's portfolio implies actually produces different results than simply chasing the hottest stocks.

What the Data Revealed

We ran both strategies through the Stax backtesting engine from June 1, 2021, through May 25, 2026, starting with $100,000 in capital. Both portfolios used equal weighting across 15 positions, monthly rebalancing, quarterly reconstitution, and realistic trading friction ($0.005 per share commission and 0.1% slippage). Both included a 20% hard stop-loss, 15% trailing stop, and 35% max drawdown circuit breaker.

The Congressional Quality strategy screened 139 symbols and completed 180 trades over 1,370 trading days. It delivered a total return of 37.4% (6.6% annualized) with a Sharpe ratio of 0.25 and a maximum drawdown of 22.6%. The win rate was 52.8% and the profit factor was 1.52. For context, a Sharpe of 0.25 means you earned a modest premium above the risk-free rate for the volatility you endured; it is livable but not exceptional. The 22.6% drawdown means your $100,000 would have dropped to about $77,400 at the worst point.

The Momentum Control screened the same 139 symbols and completed 181 trades. It returned 40.9% (7.1% annualized) with a Sharpe of 0.29 and a maximum drawdown of 23.4%. Win rate was 53.0% and profit factor was 1.55.

The momentum-heavy approach won on absolute return by about 3.5 percentage points. But look at the cost: the max drawdown was 23.4% versus 22.6%, meaning you sat through a slightly deeper trough. The Sharpe ratios were close (0.29 vs 0.25), suggesting both strategies delivered similar risk-adjusted efficiency. The real difference is in what the strategies actually held. The thesis strategy, with its 60% fundamental weight, rotated toward companies with the strongest balance sheets and cash flow profiles. The momentum control chased whatever was running. During the 2022 rate-hike correction, when growth stocks sold off hard, the fundamental-weighted portfolio recovered faster because it was holding companies whose cash flows did not depend on cheap money.

Neither strategy dramatically outperformed the other, and that itself is the finding. It suggests the filter set, the four quality metrics, did most of the work. Both strategies started with the same 139 qualifying companies. Whether you ranked them by fundamentals or momentum, you ended up with similar results because the filter already eliminated the fragile companies. The edge was in the screen, not the sort.

What You Can Do With This

The most transferable insight from this test is that the filter did the heavy lifting. When you screen for ROE above 15%, operating margins above 20%, leverage below 2x, and real cash flow per share, you have already eliminated most of the fragile, speculative, and overleveraged companies. At that point, whether you rank by momentum or fundamentals produces surprisingly similar results.

Here are three specific next steps you can run inside Stax. First, try dropping the FCF per share filter and replacing it with free cash flow yield above 3%. FCF yield normalizes by price, which means you are selecting companies where the market is undervaluing their cash generation relative to their stock price. That could tilt the strategy toward value without changing its quality character.

Second, test sector concentration. Keating bought across financials, defense, real estate, and tech. You can add a sector constraint in Stax to limit exposure to any single sector at 25% of the portfolio. This prevents the strategy from loading up on tech names during a momentum-driven rally and getting crushed when sentiment shifts.

Third, consider running the same screen with a tighter max drawdown circuit breaker, say 25% instead of 35%. During the 2022 correction, both strategies dropped into the low 80s. A 25% circuit breaker would have triggered an earlier exit and preserved more capital, though it also would have locked in losses before the 2023 recovery. That tradeoff, preserving capital versus staying invested through volatility, is worth testing explicitly.

Run The Test

Both strategy JSON files are available in the Stax public strategies directory. Run them yourself to verify:

stax backtest strategies/congressional-quality-thesis.json --start 2021-06-01 --end 2026-05-25 --capital 100000
stax backtest strategies/congressional-momentum-control.json --start 2021-06-01 --end 2026-05-25 --capital 100000