← All Articles
A magnifying glass over company filings beside twenty equal brass tiles on a dark desk.

NexusTrade · Guest article

I screened US stocks for quality. Then I put the top 20 on NexusTrade.

A $10,000 paper portfolio, a published quality screen, and a three-year test of what happens next.

Philip van den BergeFounder & CEO, IntrinsiqqOctober 6, 20268 min read

I put $10,000 of paper money into twenty US stocks selected with Intrinsiqq's quality score and a price-to-free-cash-flow filter. At the October 6 check, my Intrinsiqq Quality at a Fair Price portfolio was worth $10,364.25, a gain of 3.64%.

I run Intrinsiqq, so the screen uses our own data. NexusTrade runs the paper portfolio, records the orders, and lets anyone inspect or fork the strategy. I created it on September 23, 2026. The first recorded purchases filled on September 24.

This embed shows the current Intrinsiqq Quality at a Fair Price paper portfolio. The article's figures below were captured on October 6, 2026.
The current paper account marks, captured October 6, 2026
Account measureValue
Starting balance$10,000.00
Current marked balance$10,364.25
Change since creation+$364.25 (+3.64%)
Cash$147.82
Open holdings20 stocks
Last recorded daily balance$10,339.19 (+3.39%)

The current balance adds the latest position marks to the account's $147.82 in cash. The daily performance chart below ends at its last saved observation on October 6: $10,339.19, or +3.39%. The holdings and daily chart refresh separately.

Philip's recorded paper portfolio balanceThe daily record starts at $10,000 on September 23, reaches $9,947.01 on September 29, and ends at $10,339.19 on October 6, 2026. This chart shows the stored daily history, separate from newer holdings marks. The recorded paper balanceSeptember 23 to October 6, 2026$9,900$10,000$10,100$10,200$10,300$10,400 $10,339.19 Sep 23Sep 30Oct 6 The daily history records a 3.39% gain.
NexusTrade's saved daily history shows this paper portfolio's balance since creation. The final observation was recorded at 2:59 p.m. Central on October 6, 2026.

A few weeks of paper trading are a start. I want at least three years before judging the screen.

Follow the paper portfolioInspect the holdings, recorded performance, and plain-language strategy, or fork it into your own account.nexustrade.io

The selection

Buy good businesses at a sensible price

The idea is familiar: find good businesses, then avoid paying too much for them. It suits how I invest.

Robert Novy-Marx found that gross profitability predicted average stock returns about as well as book-to-market, and that adding profitability improved value strategies. Asness, Frazzini and Pedersen later studied quality through profitability, growth and safety. Their strategy of buying quality and shorting lower-quality stocks earned risk-adjusted returns in the US and across 24 countries.

Those studies give me a reason to test the idea. My twenty-stock, long-only basket has its own universe and thresholds, and needs its own record.

Intrinsiqq's quality score combines eight checks into a number out of 100. Its published method covers price-to-earnings, price-to-free-cash-flow, five-year growth in revenue and free cash flow, share dilution, three-year margin expansion, capital structure, and return on invested capital. The formulas, thresholds and weights are public.

I required a score of at least 80 and a positive price-to-free-cash-flow multiple of 25 or lower. For a business with positive free cash flow, a multiple of 25 corresponds to a 4% free cash flow yield.

I also required US companies filing with the SEC and a market value above $2 billion. Banks, insurers and real estate were excluded.

That last filter changed the list completely. Free cash flow is a poor comparison tool for financial businesses. My first screen was full of insurers ranked on a measure our own methodology does not use to assess them. Reinsurance companies were not what I wanted this screen to hand me.

When I ran the screen on September 16, it returned 104 companies. I took the twenty highest quality scores and gave each the same target weight: 5%. A score of 91 and a score of 88 do not give me enough reason to put three times as much money into one company.

How the Intrinsiqq screen becomes a NexusTrade portfolioUS SEC filers above two billion dollars in market value are screened for a quality score of at least 80 and positive price to free cash flow of 25 or less. Banks, insurers and real estate are excluded. Philip reported 104 matches on September 16. The twenty highest scores receive equal target weights of five percent. The same holdings are rebalanced on eligible Tuesdays in February, May, August and November after at least eighty days since the last filled buy. The screen picks the names once. NexusTrade manages their target weights. US SEC filers with market cap above $2B Banks, insurers and real estate are excluded. Quality score ≥ 80 / 100 Positive P/FCF ≤ 25 FCF yield ≥ 4% 104 matches Philip's September 16, 2026 screen 20 highest scores · 5% target each The selected companies stay in the basket. Rebalance when the portfolio is empty, or on an eligible quarterly Tuesday. February / May / August / November; ≥ 80 days since a buy. The next scheduled date is February 2, 2027.*
The 104 matches came from my September 16 screen. The selected names stay fixed while NexusTrade rebalances their weights. The next scheduled date assumes no intervening filled buy.

The selected basket contains DECK, EXEL, BKNG, INTU, LULU, TTD, RMD, PCTY, LNTH, MNST, DXCM, AGX, INSW, PRDO, PGNY, ADSK, EXPE, PTC, LOGI and MEDP.

Booking Holdings was one of the clearest examples in my original screen: a quality score of 97, operating margins near 33%, return on capital of 37.6%, and a share count about 23% smaller over the period I examined. Those are the figures from that selection, and the current company scorecard can change with prices and new filings.

Several names had fallen sharply before I selected them, including The Trade Desk, Intuit and Lululemon. Low multiples often appear where investors have soured on a business. A quality screen can keep looking wrong for a long time, and a falling price can reflect a business getting worse.

Read the Intrinsiqq scoring methodCheck the eight inputs, their weights, and the assumptions behind the score.intrinsiqq.com Inspect Booking HoldingsCompare the example above with the company’s current quality scorecard.intrinsiqq.com

The implementation

Twenty names, with a rule for when to rebalance

The portfolio holds at most twenty positions, with equal target weights and a 5% allocation cap per name at rebalance. Prices can move the weights above or below 5% between rebalances.

It buys when the portfolio holds nothing. After that, it rebalances on a Tuesday in February, May, August or November, once at least 80 days have passed since the last filled buy.

I chose those months to follow the usual US earnings cycle. This version holds the original basket, so the calendar schedules a weight reset; it does not run a fresh quality screen.

The first version tied the initial purchase to those four months too. It would have left the account in cash until November 3 while looking healthy on the dashboard. I caught the problem by reading the rule on the portfolio page. Austin fixed it the next morning.

The recorded first fills are September 24. The current position records show purchase costs of roughly $489 to $494 per stock, with $147.82 left in cash.

The 80-day condition also skips November 2026 after that September purchase. Assuming no further buys, the next scheduled rebalance is Tuesday, February 2, 2027.

The evidence

Why I skipped a backtest

A backtest using today's quality scores would give an old portfolio information it could not have had at the time.

Companies restate their financial reports. If I score a business for 2019 using figures corrected years later, the screen sees the cleaned-up version of events. The original errors may be exactly what would have fooled an investor in 2019.

There is also survivorship bias. My current company universe leaves out businesses that have disappeared through bankruptcy, acquisition or delisting. Testing an old strategy on that universe skips some of the companies it most needs to prove it would have handled.

Austin pointed out that these are limitations of the data used in this experiment. NexusTrade has filing-dated fundamentals and historical security records that can support a test using information available at the time. The query still has to use the correct filing dates and historical universe. That would be a separate experiment from backtesting my fixed basket with today's scores.

For this portfolio, I chose a forward record. The creation date, holdings and recorded orders are public on NexusTrade. I can compare future changes against the rules written down at the start.

The test ahead

How I will judge it

My intended benchmark is the S&P 500 Equal Weight Index, because this basket starts with equal weights too. I want a minimum of three years. Two bad quarters would tell me very little.

The shared dashboard's built-in comparison uses SPY, which tracks the conventional S&P 500. That is a different benchmark from the equal-weight index I plan to use for this experiment.

If the portfolio trails my benchmark over three years, I will need to question what this screen adds over a broad index approach. A short early gain cannot answer that.

The choice I am least sure about is keeping the companies fixed. If a business deteriorates, this rule keeps holding it. Re-running the screen every quarter would follow the scoring method more closely, but it would create a different strategy and require ongoing updates to the holdings.

I chose the fixed basket because it is easier to judge. The next three years will test that choice too.

What I would take from this experiment

Check what a filter does to the universe before trusting its output. Excluding financials rewrote my list. Without that step, I would have published a portfolio of insurers selected with the wrong measure.

Write down the holdings, benchmark, horizon and reasoning before the orders go through. A public record makes later changes easier to see.

When someone presents a screen through a backtest, ask which version of the financial reports it used and whether the universe includes businesses that failed or disappeared. Later restatements and missing failed companies can both make the curve measure hindsight.

Inspect or fork Intrinsiqq Quality at a Fair PriceThe portfolio page contains the holdings, the strategy, and its recorded paper performance.nexustrade.io

You can run your own screen through Intrinsiqq's screener. Company pages include quality scores and cash flow multiples, and Intrinsiqq has a free plan.

About the author

Philip van den Berge is the founder and CEO of Intrinsiqq, a stock analysis platform covering more than 10,000 global companies with quality scores, DCF valuations and ten years of financial history.

Sources

This article is for information and is not investment advice. The portfolio uses paper money. Paper results differ from real trading, and past performance does not guarantee future results.

Build the strategy this article describes

Create a free account to backtest ideas against market history, inspect the risk, and deploy to paper or live markets when you're ready.

or

Free to browse. No credit card required.

Discussion

Sign in or create a free account to join the discussion.

No comments yet.