The Chip Complex Is the Only Momentum Story Left: A 102-Stock Factor Screen (2026-09-14, Reproducible)

Most "stock pickers" on the internet can't reproduce their own picks. So here's a small, fully reproducible factor screen — real data, real methodology, real numbers — run on 102 liquid US mega-cap equities with data as of the 2026-09-11 close. You can run every line yourself and disagree with me if you want.

The model in one paragraph

Score each stock on four things, then blend:

• Momentum (12-1), 50% — return from t-252 to t-22 (we skip the last month to avoid short-term reversal).
• Quality (Sharpe), 30% — annualized mean daily return / annualized daily vol over the trailing 252 days.
• Low volatility, 10% — inverse z-score of annualized realized vol.
• Low drawdown, 10% — inverse z-score of max drawdown from peak.

All four components are z-scored across the 102-name universe before weighting. Composite = 0.50·z_mom + 0.30·z_sharpe + 0.10·z_lowvol + 0.10·z_lowdd. Prices are split/dividend-adjusted daily closes from yfinance. No black box, no "proprietary alpha."

The results (data as of 2026-09-11)

Top 10 by composite:

#	Ticker	Name	12M Ret	12-1 Mom	Sharpe	MaxDD 1Y	Score
1	MU	Micron	+597.7%	+506.2%	2.79	−39.1%	+3.659
2	INTC	Intel	+315.6%	+310.2%	2.18	−41.9%	+2.184
3	AMAT	Applied Materials	+180.9%	+223.6%	2.03	−39.6%	+1.659
4	AMD	AMD	+223.5%	+210.2%	1.99	−27.8%	+1.404
5	LRCX	Lam Research	+179.3%	+183.7%	1.92	−41.8%	+1.342
6	MPC	Marathon	+121.6%	+93.7%	2.50	−18.3%	+0.945
7	ASML	ASML	+115.5%	+126.6%	1.89	−22.0%	+0.920
8	CAT	Caterpillar	+95.3%	+100.1%	1.89	−26.7%	+0.815
9	KLAC	KLA	+94.7%	+118.0%	1.41	−43.6%	+0.776
10	FDX	FedEx	+73.8%	+79.5%	2.09	−11.7%	+0.709

Bottom 10:

#	Ticker	Name	12M Ret	12-1 Mom	Sharpe	MaxDD 1Y	Score
93	ORLY	O'Reilly	−18.4%	−14.6%	−0.66	−23.3%	−0.714
94	ZS	Zscaler	−41.0%	−38.1%	−0.50	−64.9%	−0.754
95	HD	Home Depot	−23.2%	−17.2%	−0.92	−28.8%	−0.773
96	LOW	Lowe's	−24.6%	−19.1%	−0.91	−30.8%	−0.780
97	TMUS	T-Mobile	−22.6%	−25.5%	−0.72	−29.6%	−0.784
98	COIN	Coinbase	−44.4%	−54.0%	−0.48	−63.6%	−0.908
99	NFLX	Netflix	−38.0%	−38.3%	−1.15	−45.5%	−0.938
100	ORCL	Oracle	−53.7%	−49.6%	−1.06	−64.6%	−0.980
101	HUBS	HubSpot	−53.9%	−57.7%	−0.70	−67.4%	−0.987
102	NKE	Nike	−48.8%	−44.3%	−1.67	−49.3%	−1.120

The interesting part

• 7 of the top 10 are semiconductor or semi-adjacent names. The memory cycle (Micron, Intel) and the equipment cycle (AMAT, LRCX, KLAC) are doing the heavy lifting. Micron's +598% 12-month return is the single largest contributor to its #1 score — that's a momentum regime, not a quality story, and the model is telling you exactly that.
• MPC and CAT are the "boring" top-10 names — energy and industrials with strong momentum and comparatively low volatility. They're the quality anchor of the top of the table.
• The bottom of the table is consumer discretionary + software that has underperformed: Nike, HubSpot, Oracle, Netflix, and Coinbase all carry negative 12-month momentum and negative Sharpe. That's the weakest risk-adjusted profile you can have.
• NFLX and COIN have high beta (0.72 and 2.50 respectively) with negative momentum — beta amplifies the downside score, which is exactly what a low-drawdown component is supposed to punish.

Why reproducibility matters more than the pick

Anyone can post a table. The point is that this one is reproducible: yfinance download, 252-day windows, cross-sectional z-scores, fixed 0.50/0.30/0.10/0.10 weighting. If you run it on a different date you get a different ranking — and that's the whole point of a factor model over a "hot tip."

# 1. Fetch 2y daily closes for the 102-ticker universe (yfinance)
# 2. Compute:
#    - 12-1 momentum: close[-22] / close[-252] - 1
#    - Annualized vol: std(daily_returns, 252d) * sqrt(252)
#    - Sharpe: mean(daily_returns, 252d) * 252 / annualized_vol
#    - Max drawdown: min(close / cummax(close) - 1) over 252d
# 3. Z-score each component across the universe
# 4. Composite = 0.50*z_mom + 0.30*z_sharpe + 0.10*z_lowvol + 0.10*z_lowdd
# 5. Sort descending

This is a research screen, not a recommendation. Past momentum does not guarantee future returns. All data is point-in-time and subject to revision.

---

Want the full report + the complete 102-row factor dataset (CSV)? It's published by LaunchTower, an independent market-data desk. Grab the full report and dataset here:

👉 LaunchTower — Momentum + Quality Factor Report & Dataset (2026-09-14)

(Stripe is currently in test mode — test card 4242 4242 4242 4242, any future date/CVC. Live mode activates soon; the link will collect real payments at that point.)

LaunchTower — independent market-data desk. Generated from public data; not personalized investment advice.