Blog
Factor investing and quant scoring, in plain English
Posts are organized into four clusters — pick the topic you're here for, or browse the latest below. Every post is hand-written, evergreen where possible, and cross-linked to the live product (methodology, glossary terms, ticker scores, and head-to-head comparisons).
Latest posts
- SanDisk doubled in two months. Our QScore never left HOLD — here's why.As SanDisk (SNDK) doubled over two months, its QScore barely moved off 50 — HOLD, MEDIUM confidence. Momentum at 92 saw the run; risk at 18 and below-average value and growth pulled the other way, and they net to neutral. Built entirely on our committed no-look-ahead snapshots, here's why a doubling and a HOLD belong on the same page.
- Scoring the AI IPOs: why a factor model can't read SpaceX or OpenAI yetSpaceX raised ~$75B at $135 and trades today as SPCX; OpenAI has only filed a confidential S-1. The quant question isn't "buy or not" — it's "can the model even see these yet?" Momentum and risk are undefined without price history, value is distorted by losses, and the honest output is LOW confidence. Here's the factor-by-factor read on the year's biggest listings.
- Reading the Fed like a quant: what rate decisions do to your factorsThree meetings into 2026 the Fed has held at 3.50–3.75% with dissents in both directions and one cut penciled into the dot plot. Rates are the single macro variable that most reliably reorders which factor gets paid — here's the regime-by-regime read on all five QScore factors and how to tilt without overtrading.
- DUNS is done: why the federal exit ended the gold-standard claimThe U.S. federal government quietly retired the DUNS number as its entity identifier in April 2022. The pay-to-play credit profiles, the stale trade tape, and the methodology nobody can audit had already done the damage. Here's the eulogy for the so-called gold standard.
- How the stock market behaves before and after Memorial DayPre-holiday drift is one of the oldest claims in market lore. We pulled 36 years of S&P 500 data around Memorial Day, compared it to 10,000 random 5-day windows, and the honest answer is: the direction matches the folklore, the magnitude is small, and the statistical test says "not really."
- Do stock factors actually work? Testing momentum, low volatility, and reversal on 5 years of S&P 500 dataAfter three posts on credit-side modeling, here's the equity-side test. 619,040 daily price rows. 474 names. Five years. Three factors. Momentum +3.9% annualized, low vol -6.4% (wrong sign), reversal +2.8%. The honest version of factor investing.
Market signals
- SanDisk doubled in two months. Our QScore never left HOLD — here's why.As SanDisk (SNDK) doubled over two months, its QScore barely moved off 50 — HOLD, MEDIUM confidence. Momentum at 92 saw the run; risk at 18 and below-average value and growth pulled the other way, and they net to neutral. Built entirely on our committed no-look-ahead snapshots, here's why a doubling and a HOLD belong on the same page.
- Scoring the AI IPOs: why a factor model can't read SpaceX or OpenAI yetSpaceX raised ~$75B at $135 and trades today as SPCX; OpenAI has only filed a confidential S-1. The quant question isn't "buy or not" — it's "can the model even see these yet?" Momentum and risk are undefined without price history, value is distorted by losses, and the honest output is LOW confidence. Here's the factor-by-factor read on the year's biggest listings.
- Reading the Fed like a quant: what rate decisions do to your factorsThree meetings into 2026 the Fed has held at 3.50–3.75% with dissents in both directions and one cut penciled into the dot plot. Rates are the single macro variable that most reliably reorders which factor gets paid — here's the regime-by-regime read on all five QScore factors and how to tilt without overtrading.
- How the stock market behaves before and after Memorial DayPre-holiday drift is one of the oldest claims in market lore. We pulled 36 years of S&P 500 data around Memorial Day, compared it to 10,000 random 5-day windows, and the honest answer is: the direction matches the folklore, the magnitude is small, and the statistical test says "not really."
Factor investing
- DUNS is done: why the federal exit ended the gold-standard claimThe U.S. federal government quietly retired the DUNS number as its entity identifier in April 2022. The pay-to-play credit profiles, the stale trade tape, and the methodology nobody can audit had already done the damage. Here's the eulogy for the so-called gold standard.
- Do stock factors actually work? Testing momentum, low volatility, and reversal on 5 years of S&P 500 dataAfter three posts on credit-side modeling, here's the equity-side test. 619,040 daily price rows. 474 names. Five years. Three factors. Momentum +3.9% annualized, low vol -6.4% (wrong sign), reversal +2.8%. The honest version of factor investing.
- Do stock factors work in crypto? Testing momentum, low volatility, and reversal on 1,050 tokensAfter testing factors on 5 years of S&P 500 data and finding nothing significant, we pointed the same machinery at 1,050 cryptos over 6 years. Momentum IC went from +0.016 to +0.111 (t-stat +3.80). Low-vol flipped sign. Reversal flipped sign and lost 20%/year. The honest crypto version of factor investing.
- How credit scoring models actually work: a data-driven breakdownMost explanations of credit scoring stop at "lenders look at your income and credit history." That's not wrong — it's just not useful. So we pulled 32,437 real loan applications, trained a working model, and looked at what the numbers actually say.
- Predicting loan defaults: what the data tells us banks missThe previous credit-risk dataset was generous: AUC 0.87 with simple logistic regression. This one isn't. 67,463 loans, 35 features, and even random forest gets to AUC 0.527 — barely above random. Data quality beats model choice, every single time.
- Detecting credit card fraud: when 99.8% accuracy means your model caught nothingA model that predicts "not fraud" for every transaction in this dataset is right 99.83% of the time and catches zero fraud. We trained two real fraud models on 284,807 transactions and looked at which evaluation metric actually tells the truth.
Stock metrics
- P/E ratio explained: how to read price-to-earnings (with real ticker examples)Every stock-screener tool ranks P/E. Most readers see the number without knowing exactly what's in the numerator and denominator — or why a low P/E isn't always cheap. Here's the plain-English breakdown.
- RSI explained: how to read the Relative Strength Index (and where it fails)RSI is the workhorse momentum oscillator — overbought above 70, oversold below 30. The reality is more nuanced: extreme readings can persist for weeks, and the indicator fails most spectacularly at regime turns.
- Beta explained: what it measures, how it's computed, and why it can misleadBeta has been finance education's go-to risk metric for sixty years. The textbook story is clean — and the empirical story is much messier. Here's what beta actually measures, where it works, and why high-beta stocks haven't paid off the way CAPM predicted.
- Sharpe ratio explained: the most-cited measure of risk-adjusted returnEvery quant strategy you'll ever read about reports its Sharpe ratio. The metric itself is decades old and well-defined; reading it well takes more nuance than 'higher is better.'
QScore methodology
- How to read a QScore: the five factors explainedEvery QScore is a weighted average of five factor categories, each scored from 0 to 100 against the ticker's sector. Here's what each one is actually measuring, why it's there, and how to read the breakdown when the composite alone isn't enough.
- What is the QScore? A transparent quant signal for any US stockThe QScore distills decades of factor-investing research into a single number with a clear signal. Here's what it is, how it's built, what it isn't, and the validation pledge that keeps the methodology honest.
Stock comparisons
- NVDA vs AMD: how the QScore breakdown reveals two different AI betsSame sector, similar narrative, very different factor profiles. Here's what the QScore breakdown reveals about NVDA vs AMD — and why treating them as interchangeable AI plays is the most common mistake.
- AAPL vs MSFT: which megacap looks better on the quant scorecard?Same size, same index weight, similar composite QScores — and very different bets underneath. Here's how AAPL and MSFT diverge on the factor breakdown, and what the typical pattern means for what you're actually owning.
- GOOGL vs META: ad duopoly stocks under the QScore lensBoth are ad-revenue platforms. Both ride the same digital-ad tailwind. Their composite QScores can land close — but the factor signatures show very different exposure profiles. Here's how to read the comparison.