Quantarded Weekly Signals #009 — Week 8, 2026

This week closed negative, with dispersion again concentrated in a handful of names.

The basket saw an early drawdown, a partial mid-week recovery, and a modest fade into Friday. The final print was close to flat on a weighted basis, but the path once again highlighted how sensitive concentrated baskets are to a single underperformer.

For context, this week Quantarded processed 184 House trade disclosures filed during the week, 227,058 Reddit comments analyzed and 23,495 stock ticker mentions detected and classified. As always, what matters is not volume alone, but how agreement evolves as volume grows.

Reddit picks

As a reminder, a ticker is labeled BUY or SELL only when it clears a minimum imbalance threshold. High visibility alone is not sufficient; divided sentiment is explicitly penalized.

This week’s basket is structurally different from last week: it includes two large mega-caps, one dominant AI infrastructure anchor, and two SELL signals that clear the threshold with meaningful confidence. In practice, that makes the basket feel more "macro beta adjacent" than a pure meme-week, even though the signals are still driven by imbalance, not popularity.

$NVDA: BUY, 30% share

$NVDA is the top weight this week, with the highest confidence in the basket and a clear positive imbalance.

This looks like sustained directional agreement rather than a one-off comment spike. Breadth is also present: our weekly analysis shows high participation across both submissions and comments, which typically makes the signal more stable than low-volume, high-imbalance names.

Nvidia is scheduled to report 4th Quarter FY26 results on February 25, 2026: NVIDIA Sets Conference Call for Fourth-Quarter Financial Results.

Additionally, Nvidia has been in the news cycle around continued hyperscaler demand for AI hardware, including a reported multi-year chip supply agreement: Nvidia to sell Meta millions of chips in multiyear deal.

$MSFT: SELL, 23% share

$MSFT is the second-largest position but the dominant SELL signal, with high confidence and a clear negative direction.

This is not a "hate ticker" effect. The model is seeing enough imbalance to clear the threshold, with participation broad enough that it is unlikely to be driven by a single thread. Structurally, this makes $MSFT one of the key drivers of week-to-week performance because its share is large and the direction is explicitly negative.

As contextual background only, Microsoft has had prominent coverage related to the scale of its AI and data-center buildout and the energy implications of that expansion: Microsoft exploring using advanced power lines to make data centers more energy efficient and Microsoft to keep buying enough renewable energy to match all its electricity needs.

$CVNA: SELL, 19% share

$CVNA is a mid-weight SELL with moderate confidence but a strong imbalance profile relative to its size.

This is the kind of signal that tends to be more path-dependent. The conviction clears the threshold, but it is less "baseline stable" than the top two names because participation is lower and idiosyncratic volatility tends to dominate outcomes in a single-name used-auto retailer.

Carvana was recently covered around earnings and cost pressure: Carvana shares tumble as fourth-quarter profit misses on higher costs and Used-car retailer Carvana shares tumble as vehicle reconditioning costs rise.

$AMZN: BUY, 17% share

$AMZN enters as a lower-weight BUY with moderate confidence, but a high imbalance characteristic in the internal report.

That usually implies a cleaner directional skew among participants, even if total participation is not as dominant as the top names. In basket terms, this adds breadth, but it does not carry the portfolio unless the top weights are neutral.

Amazon has been covered on capital spending and AWS-related operational themes: Amazon sees 50% boost to capital spending this year, shares tumble and Amazon's cloud unit hit by outage involving AI tools.

$AAPL: BUY, 11% share

$AAPL is the smallest position and the lowest-confidence inclusion, but it still clears the imbalance requirement.

This is very much a "threshold BUY": it contributes diversification but not stability. If the week turns into a regime move where the portfolio is decided by one or two names, $AAPL is unlikely to be the deciding factor at this weight.

Apple has been in the news cycle around demand, outlook, and supply-chain cost pressures: Apple forecasts strong sales growth as iPhone demand rebounds and Pricier iPhones? Global memory chip crunch puts spotlight on Apple.

Also, Apple has invited press to a “special Apple Experience” on March 4, 2026, where new hardware announcements are expected: Apple special event announced for March 4.

Note: several highly visible tickers appeared heavily discussed in the internal report but did not clear the imbalance threshold for inclusion. Visibility without conviction is not a signal.

House trades

House trades produced one qualifying pick in the Week 8 House dataset:

• BUY "United States Treasury Notes 3.5% 31-JAN-2028" (~$100,001), disclosed by Richard W. Allen (traded 2026-01-26, filed 2026-02-17, disclosed 2026-02-18)

The rest of the disclosures feed is much busier, but most of it does not clear the sizing filter. A few factual notes from what did not make the cut:

• The feed is dominated by one filer: Gilbert Cisneros accounts for 140 of 194 disclosures in the weekly report.
• The vast majority of trades are the minimum bucket: 175 of 194 disclosures are in the $1,001 range, which is structurally uninformative for a “meaningful notional” screen.
• Even where tickers repeat, it is shallow: the most repeated tickers appear only 3 times (notably $DASH and $NVDA), with most others repeating 2 times or less.
• Directionally, the complete feed leans BUY: 121 vs 73 SELL disclosures.

Performance review

Last week’s results

Early weakness (particularly Tuesday) drove the weekly drawdown. Mid-week strength was visible in several names, but Friday pressure in the more volatile components limited recovery.

Dispersion was again the defining feature. Two semiconductor names finished solidly positive, one crypto-linked name detracted, and one EV name posted a double-digit weekly loss.

This was a week where concentration mattered. The basket’s weighted result was significantly more stable than the simple average, underscoring how top-weight positioning shaped the final outcome.

Portfolio tracking

• End of 26W7 return: -0.32%
• YTD (2026) return: +13.05%
• Cumulative return since inception: +24.55%

As of 2026-02-22, the portfolio stands at $12,454.96 starting from $10,000 on 2025-12-21.

Knowing the algorithm: breadth vs concentration

When a ticker makes the weekly basket, there are two very different ways it can get there.

The first is breadth: lots of independent events across the week contribute to the same direction. Many different comments, several different submissions, repeated mentions that survive the recency decay. This tends to read like “the crowd kept agreeing”, not “the crowd briefly noticed.” In our model, breadth matters because the score is built by aggregating many weighted events over the ISO week, not by spotlighting a single viral moment.

The second is concentration: most of the week’s signal comes from a small number of events, often anchored by one submission that generates a long comment cascade. That can still clear the imbalance threshold and become a valid signal, but it is usually more fragile. If the conversation stops, or if later-week discussion rotates to a different idea, concentrated signals can fade faster because the model is recency-weighted and continuously recomputed as the week progresses.

This is why we keep an internal “extended” view alongside the final picks. The weekly output is one number per ticker (sentiment + confidence), but the extended breakdown is what tells you whether the signal is supported by many distinct events or just a narrow burst. It is the difference between a choir and a soloist.

A simple way to interpret it, without inventing any new metrics:

• If a ticker’s score is supported by lots of distinct events (not just one hot thread), treat it as more stable.
• If it is dominated by a small number of events, treat it as more path-dependent, even if the headline confidence is high.

None of this predicts price. It is just signal hygiene: understanding whether the model is seeing repeated agreement across the week, or a concentrated spike that happened to be directional.

Disclaimer

This newsletter is not financial advice.

All content is provided for informational and educational purposes only. Markets involve risk, including loss of principal. Past performance does not guarantee future results. Always do your own research.

Links

Website: https://quantarded.com

Subscribe: https://quantarded.kit.com/products/monthly-subscription?step=checkout