Learn Framler in plain English

Framler score is a 0–100 composite number summarising how attractive a stock looks across 13 academic factor families — quality, value, momentum, post-earnings drift, accruals, insider flow, and others. 50 means neutral; above 60 is the buy bucket; below 40 is the sell bucket. The number is relative to the live universe, not absolute.

Each of the 13 factor families (quality, value, momentum, etc.) scores the stock 0-100. We z-score them across the universe, weight each by its prior reliability, and combine into one composite. 50 = neutral / average. 65+ = strong-buy region (top decile). Below 35 = strong-sell region. The number does not say "this stock will go up by X%" — it says "this stock currently looks more attractive than this percent of comparable names".

A verdict is the plain-English bucket Framler assigns to each stock based on its Framler score: BUY when the score is 60 or above, MIXED for scores between 40 and 60, and SELL when the score drops to 40 or below. Three buckets are used deliberately to avoid false precision.

BUY = score ≥ 60 (factor stack agrees stock is attractive). MIXED = 40-60 (factors disagree, no clear edge). SELL = ≤ 40. We deliberately use 3 buckets, not 5 or 10 — finer granularity gives false precision when forward returns aren't calibrated yet.

A forward return is the model's best estimate of how much a stock will return over the next 1, 7, 30, or 90 days, expressed as a mean plus a 90% confidence band. The engine composes the estimate from a market-drift baseline and the per-ticker alpha implied by the composite score, then adjusts for the current regime, any matching academic pattern, and tail-dependence shrinkage. Wider bands signal lower confidence, not lower expected return.

Computed by the engine v10 Forward-Return Engine. The mean projection combines a market-drift baseline (small, slowly drifting) with the alpha implied by the score, then is shaped by the current market regime and any matching confluence pattern. We show 4 horizons (1d, 7d, 30d, 90d) with mean, 90% CI, and P(positive return). Prior-mode until 2026-05-16 — read it as "engine's theoretical projection from published factor literature", not "guaranteed outcome". The exact composition formula and its calibrated coefficients are proprietary.

A market regime is a probabilistic label for the current mood of the overall market: risk-on (broad bullish flow), risk-off (broad bearish flow), or transition (uncertain). Framler uses Bayesian Online Changepoint Detection on SPY log-returns to maintain a live posterior over the three regimes, updated daily.

Adams-MacKay 2007 Bayesian Online Changepoint Detection — we maintain a posterior distribution over which regime the market is currently in, updated daily on SPY log-returns. Each factor has different efficacy in each regime (e.g. momentum works great in risk-on, gets crushed in risk-off). The regime modulates how we weight factors and how we amplify the alpha contribution.

A conformal prediction interval is a confidence band with a finite-sample, distribution-free coverage guarantee. If the engine reports a 90% conformal interval of [56, 82], then 90% of similar future predictions will contain the realised value — assuming exchangeability. Wider intervals reflect higher per-ticker uncertainty, not lower expected return.

Vovk-Gammerman-Shafer 2005 conformal prediction provides finite-sample, distribution-free coverage guarantees: if the model says [56, 82] is a 90% interval, then 90% of similar future predictions WILL contain the realised value. We use Mondrian conformal (different intervals per regime + verdict bin). Wider interval = more uncertainty about this name.

How often a pattern's factors actually move together in extreme conditions, not just on average.

Schmidt-Stadtmüller 2006: an empirical estimator for whether two factors co-move in their tails (extreme upper or lower decile). λ ≥ 0.25 means the factors historically co-move when the market is stressed — structural pattern, not coincidence. λ < 0.10 means they're independent in tails — the apparent setup is probably random alignment. We use this to dampen pattern confidence when tail-alignment is weak.

A way to update factor weights smoothly over time as their informational content drifts, rather than re-fitting from scratch.

Kalman 1960 / West-Harrison 1997 dynamic linear model: each factor's coefficient on forward returns is treated as a slowly-evolving random walk. We update the posterior on each new observation. This avoids the choice between "use the prior forever" (rigid) and "re-fit on a rolling window" (jumpy). Activates as forward returns accumulate post-2026-05-16.

A fair way to split the credit/blame for the score among the 13 factors — each factor gets exactly its marginal contribution.

Shapley 1953 game-theoretic value: across all permutations of factor inclusion, what's factor X's average marginal contribution to the final composite? This gives a cleaner attribution than simple "factor i has weight w_i × score" because it accounts for factor interactions. The Score Composition waterfall on every ticker page shows Shapley contributions.

A named setup the engine recognises when multiple factors align — like "Quality Compounder" or "Short Squeeze".

When a specific combination of factor scores fires, we label it with one of 19 academic-style patterns (Quality Compounder, Value Trap, Short Squeeze Setup, Earnings Quality Crack, etc.). Each pattern has a published source effect size. The pattern overrides part of the raw composite — that's the +δ shown on the banner. Tail-dependence determines whether the override is full-weight (structural co-movement) or half-weight (coincidence).

A z-score is a number that tells you how many standard deviations a value sits away from the mean of its distribution. A z-score of +2 means the value is in the top 2.5% of the distribution; −2 means the bottom 2.5%. Framler z-scores every factor across the live universe so a 70 means top-quintile relative to today's peers.

Formally z = (x − μ) / σ. Z-scoring is the standard normalisation that lets us compare value (whose raw scale runs from 0 to infinity) against momentum (which runs negative to positive). After z-scoring, every factor lives on the same axis and can be combined into a composite without one factor dominating purely because its raw scale is larger.

A factor model is a framework that decomposes a stock's return into exposures to a small number of systematic drivers (factors) plus an idiosyncratic component. The CAPM has one factor (market beta); Fama-French has three (market, size, value); modern multi-factor models like APEX cover quality, momentum, accruals, sentiment, and others.

Mathematically: r_i = α_i + Σ β_ij · F_j + ε_i, where r_i is stock i's return, F_j is factor j's return, β_ij is stock i's exposure to factor j, and ε_i is idiosyncratic noise. Factor models let us rank stocks cross-sectionally, decompose performance into compensated and uncompensated risk, and identify whether observed alpha is genuinely uncorrelated or just an unmeasured factor exposure.

Companies that consistently make money on what they own — high gross profit per dollar of assets.

Novy-Marx 2013 showed that gross profitability (revenue − cost of goods sold, divided by total assets) predicts cross-sectional returns better than book-to-market value alone. High-quality firms compound; low-quality firms grind. We score 0-100 with 100 = highest GPA in our universe.

Stocks that look cheap relative to their fundamentals (low P/E, low P/B, etc).

Fama-French 1992 famously documented that high book-to-market stocks ("value") outperform low book-to-market stocks ("growth") in long horizons. We blend P/E, P/B, EV/EBITDA, and forward-PE divergence into one z-scored value factor. 100 = cheapest in our universe by composite, 0 = most expensive.

Stocks that have been going up recently tend to keep going up; the ones falling tend to keep falling — for medium horizons.

Jegadeesh-Titman 1993: 12-month-minus-1-month price return is one of the most replicated anomalies in finance. We use 12m−1m return + revenue acceleration. The "minus 1m" part avoids the well-known short-term reversal effect (Jegadeesh 1990). High momentum = recent winner. Low momentum = recent loser.

PEAD — Post-Earnings Announcement Drift — is the empirical pattern that stocks which beat their earnings consensus keep drifting upward for 30 to 90 trading days after the announcement, while stocks that miss keep drifting down. The market under-reacts to the surprise; the drift fills in the gap as institutional flow catches up.

Bernard-Thomas 1989 documented that stocks under-react to earnings surprises — the price moves on the day, but more drift continues for 60+ days. We measure standardized unexpected earnings (SUE) and recent earnings revisions to capture this. PEAD is one of the longest-standing anomalies in finance, persistent through decades of arbitrage attention.

The accruals signal measures how much of a company's reported earnings came from accounting estimates instead of cash. High-accrual earnings tend to reverse the following year — receivables uncollected, inventory written down — so high-accrual stocks systematically underperform. Sloan 1996 documented a 10% annual return spread between low- and high-accrual deciles.

Sloan 1996: investors fixate on reported earnings without distinguishing the cash component from the accrual component. Firms with high accruals (lots of "earnings" that haven't turned into cash yet) systematically underperform. We compute working-capital accruals scaled by total assets — high accruals = bearish signal. This factor is INVERTED in scoring (lower accruals = higher score).

Stocks heavily shorted by institutions tend to underperform — the shorts are usually right on average.

Asquith-Pathak-Ritter 2005: short interest above 5% of float predicts negative future returns. But there's a flip side — extreme crowded shorts can squeeze higher. We use short-interest-to-float ratio + days-to-cover. Score is INVERTED: low short interest = high score (bullish), high short interest = low score (bearish). The "SHORT_SQUEEZE_SETUP" pattern flags potential squeezes.

When company executives buy their own stock with own money, that's a strong bullish signal. Selling is weaker as a bearish signal (insiders sell for many non-information reasons).

Seyhun 1998 (the canonical text on insider trading): open-market purchases by officers and directors predict positive abnormal returns. We pull SEC Form 4 filings, weight by transaction size, scale by company market cap. Score uses net dollar flow over rolling 90-day window — insider purchases push score up, sales push it down.

Smart-money option-buying shows up before stock-price moves. Heavy call buying = bullish lead; heavy put buying = bearish.

Pan-Poteshman 2006: option order flow contains private information that hasn't reached the equity market yet. We track put/call ratio (volume) + IV skew. Low P/C ratio with high call demand = bullish. High P/C ratio = bearish.

When a key supplier or customer makes news, the linked company's stock moves a few days later — investors are slow to update.

Cohen-Frazzini 2008: economic links between firms (supplier-customer relationships) predict returns because the market fails to instantly reflect news from one firm into the price of its linked firm. We track sector-peer momentum and industry-peer earnings surprises as proxy for these links. High peer momentum that the focus stock hasn't yet absorbed = bullish spillover.

Reading the company's 10-K filing for negative-sounding words — companies whose MD&A is unusually negative tend to underperform.

Loughran-McDonald 2011 built the standard financial-sentiment dictionary (after showing that general English sentiment dictionaries fail catastrophically on financial text — "liability" is not a negative word in finance). Li 2008 showed that 10-K MD&A negativity has predictive power. We score the tone of the most recent 10-K MD&A section using LM-2018 dictionary; high negativity = bearish.

Kelly sizing is the formula for how much capital to allocate to a high-confidence position to maximise long-run compound growth without risk of ruin. Full Kelly is too aggressive in practice — most practitioners use ½-Kelly, which gives up a small amount of expected growth for substantially lower variance and drawdown risk.

Kelly 1956 derived the bet size that maximises long-run log wealth. In practice, full Kelly is too aggressive — a single mis-estimate of edge tanks the portfolio — so practitioners use ½-Kelly. Framler infers edge from the composite’s divergence from neutral, modulated by interval width and conviction, then caps the recommendation at a conservative fraction of portfolio so a single high-conviction call cannot blow up the account. Output: dollars per $10k allocated.

The Sharpe ratio is the annualised excess return of a strategy divided by its annualised volatility. A Sharpe of 1.0 is solid, 2.0 is institutional-grade, and above 3.0 is rare and often suggests either short backtest length or hidden leverage. The metric measures reward per unit of risk, not absolute return.

Defined by William Sharpe 1966 as (R_p − R_f) / σ_p, where R_p is portfolio return, R_f is the risk-free rate, and σ_p is the standard deviation of excess returns. The intuition: a 30% return with 60% volatility (Sharpe 0.5) is worse than a 10% return with 5% volatility (Sharpe 2.0) because the second strategy compounds more reliably. APEX targets Sharpe in the 1.0-1.5 range across realised forward returns post-2026-05-16.

Max drawdown is the largest peak-to-trough decline a strategy or stock has experienced over a measurement window. A 30% max drawdown means at the worst point the position was down 30% from its prior high. The metric matters because most retail investors abandon strategies during drawdowns, locking in losses.

Computed as max over all (t1, t2) of (peak[t1] − trough[t2]) / peak[t1] for trough[t2] occurring after peak[t1]. Strategies with high Sharpe but very deep drawdowns are psychologically hard to hold — Long-Term Capital Management had Sharpe near 4 right before its 2-week 90% drawdown. APEX position-sizing via Kelly and conformal intervals is designed to keep individual-position drawdowns inside tolerable ranges.

Alpha is the portion of a strategy's return that cannot be explained by exposure to known factors like the market, value, momentum, or quality. Positive alpha means the strategy beats what you'd expect from its factor loadings alone — true skill or edge. Most retail strategies have zero or negative alpha after fees.

Originally defined by Jensen 1968 as the intercept term in a regression of strategy returns against a factor model (e.g. CAPM, Fama-French 3-factor, or Carhart 4-factor). A 5% annualised alpha means the strategy delivered 5% per year that no exposure to standard factors can account for. Hard to achieve, easy to fake via short backtest or transaction-cost neglect — APEX measures alpha against a multi-factor stack with realistic costs.

Phase 3 failure probability is the model's estimate of the chance that an active Phase 3 clinical trial will miss its primary endpoint at readout. The base rate per Hay et al. 2014 is roughly 42%; Framler Bayesian-updates this with enrollment velocity, endpoint amendments, mechanism evidence from PubMed, and sponsor cash runway.

Hay et al. 2014 showed Phase 3 trial success rates anchor around 58% (so failure rate ~42% as a base). We Bayesian-update this base rate with live signals: enrollment velocity, endpoint amendments, mechanism evidence from PubMed, sponsor cash runway. Output: P(failure) per active Phase 3 + days to readout. Phase 3 failures move stocks 30-50% overnight, so timing matters as much as probability.

Catalyst days is the count of trading days until a binary event for a pharma or biotech ticker — Phase 3 trial readout, FDA approval decision, or major regulatory ruling — that historically moves the stock 30% or more in a single session. Under 90 days is imminent; over 365 is distant enough that current pricing already reflects most of the risk.

Pulled from ClinicalTrials.gov primary completion date for active Phase 3 trials. <90 days = imminent (size positions smaller). 90-365 = mid-term. >365 = distant (binary risk far enough that current pricing already reflects most of it).