I Ranked 151 US Large-Caps With a 15-Line Momentum + Quality Model — Here's the Full Code

Free dataset, full methodology, and a reproducible script. No paywall on the data — just a $29 pack if you want it delivered, dated, and documented.

---

Why I built this

I keep seeing "factor investing" posts that show a pretty chart and a conclusion, but never the code. You can't verify what you can't run.

So I built a small, honest, reproducible factor screen on 151 liquid US large-caps and published everything: the data, the report, and the exact Python that produced it. If you can run the script, you can trust the numbers.

---

The model (it's genuinely simple)

Five numbers per ticker, computed from price data only:

1. 12-month momentum — return from 252 trading days ago to 21 days ago (skipping the last month to dodge short-term reversal)
2. 3-month momentum — return from 63 days ago to 21 days ago
3. Volatility — 252-day annualized standard deviation of daily returns
4. Quality proxy — inverse of volatility (lower vol = higher quality)
5. Composite score — 50% momentum z-score + 50% quality z-score

That's it. No black box, no look-ahead bias, no "proprietary alpha." Just momentum and quality, which are two of the most robust anomalies in the academic literature.

---

What the screen says (2026-09-16)

Top 5 by composite score:

Rank	Ticker	12M Momentum	3M Momentum	Volatility	Composite
1	APD	+6.3%	-9.6%	46.8%	+6.07
2	PBR	+35.3%	+18.1%	28.8%	+2.80
3	AVGO	+30.9%	-1.4%	39.2%	+2.35
4	NVDA	+45.9%	+37.6%	34.8%	+1.96
5	TRV	-11.5%	+5.5%	46.0%	+1.86

Bottom 5:

Rank	Ticker	12M Momentum	3M Momentum	Volatility	Composite
151	SNOW	-2.6%	-9.7%	51.9%	-1.91
150	CME	-40.5%	-8.7%	35.9%	-1.68
149	ETN	-13.0%	+6.8%	25.1%	-1.66
148	META	-22.9%	+21.4%	37.0%	-1.61
147	GILD	-15.2%	+5.3%	25.3%	-1.60

What stands out:

• APD (Air Products) leads on quality alone — a +11.06 quality z-score despite negative 3M momentum. Low volatility is doing heavy lifting.
• NVDA has the strongest momentum story (+45.9% 12M, +37.6% 3M) but average quality. The AI trade is still alive, just not the cheapest.
• SNOW ranks last: high vol (51.9%) and negative momentum. The cloud software trade is under pressure.
• CME shows the weakest 12-month momentum (-40.5%) — a significant drawdown.
• META is in transition: positive 3M momentum (+21.4%) but weak 12M (-22.9%) and high vol.

---

The full code (run it yourself)

import yfinance as yf
import pandas as pd
import numpy as np
from datetime import datetime

universe = ["AAPL","MSFT","NVDA","GOOGL","AMZN","META","TSLA","AVGO","AMD","NFLX",
"ORCL","CRM","ADBE","CSCO","QCOM","TXN","MU","INTC","IBM","NOW","INTU","PLTR",
"SNOW","DDOG","NET","CRWD","PANW","ZS","FTNT","ANET","SMCI","ARM","MRVL","LRCX",
"AMAT","KLAC","ASML","ON","MPWR","MCHP","TER","ADSK","CDNS","SNPS","GFS","MRNA",
"LLY","NVO","UNH","JNJ","PFE","MRK","ABBV","BMY","TMO","DHR","ISRG","VRTX","REGN",
"AMGN","GILD","BSX","CVS","CI","HUM","ABT","SYK","ALGN","MDT","BABA","JD","PDD",
"SE","BIDU","UBER","ABNB","DASH","COIN","HOOD","PYPL","V","MA","AXP","BLK","SCHW",
"C","BAC","WFC","JPM","GS","MS","SPGI","ICE","CME","MCO","AIG","MET","PRU","TRV",
"ALL","CB","PGR","SPOT","T","VZ","TMUS","CMCSA","DIS","WMT","COST","HD","MCD","NKE",
"SBUX","TGT","UPS","CAT","DE","GE","BA","HON","UNP","CSX","NSC","LIN","APD","ECL",
"SHW","EMR","ETN","PH","ROK","WM","RSG","COP","XOM","CVX","SLB","OXY","EOG","DVN",
"PSX","VLO","MPC","PBR","BP","SHEL","TTE","RIO","FCX","NEM"]

end = datetime(2026, 9, 16)
start = datetime(2024, 9, 16)

data = yf.download(universe, start=start.strftime('%Y-%m-%d'), end=end.strftime('%Y-%m-%d'),
                   auto_adjust=True, progress=False, threads=True)

close = data['Close']
rets = close.pct_change(fill_method=None)

mom_12m = close.shift(21) / close.shift(252) - 1
mom_3m = close.shift(21) / close.shift(63) - 1
vol = rets.rolling(252).std() * np.sqrt(252)
quality = -vol

mom_z = (mom_12m - mom_12m.mean()) / mom_12m.std()
qual_z = (quality - quality.mean()) / quality.std()
composite = 0.5 * mom_z + 0.5 * qual_z

results = pd.DataFrame({
    'ticker': universe,
    'momentum_12m': mom_12m.iloc[-1].values,
    'momentum_3m': mom_3m.iloc[-1].values,
    'volatility': vol.iloc[-1].values,
    'quality_score': qual_z.iloc[-1].values,
    'composite_score': composite.iloc[-1].values
}).sort_values('composite_score', ascending=False).reset_index(drop=True)

print(results.head(10))
print(results.tail(10))

Dependencies:

pip install yfinance pandas numpy

---

What's free vs. what's in the pack

Free (no signup, no email):

• GitHub repo with the full CSV dataset — 151 tickers, all columns, dated
• Full research report — methodology, top/bottom tables, factor distributions
• The code above — run it, verify it, modify it

The $29 pack (if you want it delivered, dated, and documented):

• The same CSV, delivered to your inbox on purchase
• The full research report (Markdown)
• The reproduction script
• A dated, versioned snapshot you can cite in your own work
• No subscription, no recurring fee — one-time purchase

Get the pack: LaunchTower Momentum Signal — $29

Store: whop.com/biz_PafLwqOjrf2HRB

The data is free. The pack is for people who want a clean, dated, documented artifact they can drop into a notebook or a report without re-running the pipeline.

---

What this is NOT

• Not investment advice. I'm a research desk, not a financial advisor.
• Not a prediction. Factor models describe past structure; they don't guarantee future returns.
• Not a signal to buy or sell. It's a ranking. What you do with it is your call.

Past factor performance does not guarantee future results. Do your own due diligence.

---

Why I publish the code

Because "trust me" is not a methodology. If you can run the script and get the same numbers, you can trust the model. If you can't, you shouldn't.

That's the whole point.

---

Report generated: 2026-09-16
Data as of: 2026-09-11 (most recent trading day)
Next update: Daily

LaunchTower — independent market-data research. Not investment advice.