Why I Built FactorForge
April 2026
I've been a long-term, buy-and-hold investor for years — the kind who reads annual reports for fun and loses sleep over debt-to-equity ratios, not stock prices. My holdings are spread across Vanguard, Fidelity, and Robinhood. Every brokerage shows you your own slice. None of them tell you what your portfolio actually says about you as an investor.
That question nagged at me: Am I a value investor? A growth investor? Something in between? I had some ideas, but not the data analysis to prove it.
The problem with brokerage apps
Every brokerage gives you a view of your own account — your returns, your allocations, maybe a pie chart of sectors. But none of them synthesize across accounts, and none of them answer the more interesting question: what investment philosophy does this portfolio reflect?
I had a rough sense of what I owned. I had no sense of what it meant. Was my collection of holdings coherent? Was I inadvertently doubling down on the same factor bets across accounts without realizing it?
The tool gap
The tools that could answer that question fell into two camps. The simple ones — personal finance apps, brokerage dashboards — gave me returns and pie charts. Useful, but not what I was looking for. The sophisticated ones — Bloomberg, FactSet, institutional-grade factor platforms — were built for professional portfolio managers and priced accordingly.
There was nothing in between: rigorous enough to be meaningful, but built for a regular person who just wants to understand what they own.
What I built
FactorForge analyzes a portfolio through six evidence-based investment lenses — the same frameworks that institutional factor investors use to categorize and evaluate holdings:
- Value — stocks trading below their fundamental worth
- Growth — companies with above-average earnings expansion
- Quality — financially strong businesses with stable earnings
- Momentum — recent price and earnings trend strength
- GARP — growth at a reasonable price
- Balanced — equal weight across the other five lenses
Each holding is scored against all six lenses using fundamental data from SEC EDGAR and market data. Scores are sector-calibrated — a P/E ratio that signals value in utilities means something different in technology. The result is a per-holding verdict for each lens, rolled up to a portfolio-level picture.
The goal: a clear answer to "What kind of investor does my portfolio actually think I am?"
My portfolio, analyzed
So I ran my own portfolio through it. 48 holdings, spread across Vanguard and Fidelity funds, a handful of individual stocks held at Robinhood, and a small position in bonds. Five of six lenses scored "Strong" (the highest rating). The exception was Value, which came in at "Moderate" — the portfolio's clearest signal, and the one I least expected.
The observation the model generated: "this portfolio reads most like a growth portfolio and least like a value portfolio."
That stung a little. I think of myself as a fundamentals-first investor. But the data told a different story.
The reason is mostly structural. My largest holding by far is a Fidelity target-date fund — it's 64% of my total portfolio value. Rather than scoring that fund on aggregate metrics like its 3-year return or beta, FactorForge looks through it to the underlying equities — tracing through the fund's sub-funds all the way down to the individual stocks inside — and scores from there.
Same for the Vanguard target-date and balanced funds that make up the rest of the core. The portfolio is really 72% "fund core with individual stock exposure underneath." When you look at what those funds actually hold, the result is a heavily diversified global equity portfolio with a structural tilt toward technology and growth names — which is what the market has rewarded for years. Growth and momentum dominate because the underlying holdings do, not because of how any individual fund has performed recently. Momentum actually scores highest numerically — but momentum tells you how a portfolio has moved recently, not why. The model reserves its "reads most like" label for the fundamental lenses — and on those, Growth leads.
The individual stock picks I've added over the years — NVDA, AAPL, AMZN, TSLA, CRWD, NFLX, and about 30 others — collectively amount to just 1.4% of the total portfolio. They're the part I think about most, but they barely move the aggregate needle. The portfolio is, in practice, a broadly diversified index-fund core with a satellite layer of growth-tilted individual bets that are too small to change the overall character.
The surprise was Tesla. TSLA scores negative on the growth lens — the model sees it as a growth detractor based on its current fundamentals, despite the common perception of it as a high-growth name. A few of the smaller, more speculative positions scored only "Moderate" across the board, which tracks — these are higher-risk picks without the earnings profile to back a strong factor score.
What does this mean? The portfolio reads as growth-leaning because the core is a diversified slice of the U.S. and global equity market, and the companies inside that market have been growth- and momentum-driven for years. My individual picks reinforce that lean — I've been drawn to technology and innovation names — but they're not the reason for the score. The index funds are.
Whether that's a good thing is a separate question — one the tool deliberately doesn't answer. The goal is clarity, not advice. And on that front, it delivered: I now know exactly what kind of investor my portfolio thinks I am, even if it's not quite the story I was telling myself.
What's coming next
FactorForge is actively developed. On the roadmap: a what-if simulator to see how adding or removing a position changes your factor profile, regime sensitivity analysis, and saved portfolios with tracking over time.
If you'd like to follow along as the tool evolves, sign up for occasional updates — low frequency, no marketing, just product news.