Settings
loading saved valuesConfigure the scoring model and the portfolio rules. Changes are saved explicitly and then used by Dashboard, Rankings, Companies, Portfolio and Research.
Deterministic ranking output from public SEC XBRL filings and IBKR prices. For research and education only. Nothing here is a recommendation to buy or sell any security.
Backend
Settings are stored in this browser and apply as soon as a request succeeds. They do not depend on the API being up.
API base https://snapshot
Scoring weights
How much each category contributes to the overall model score. The five must total 100.
The five category weights must sum to exactly 100 before the model configuration can be saved. This is deliberate: a total other than 100 would make the overall score incomparable between runs.
A category that cannot be computed is never scored as zero. It renders as N/A, drops out of the weighted mean, and the remaining category weights renormalize over what is available.
Saved settings are read through lib/settings and applied on Dashboard, Rankings, Companies, Portfolio and Research. Rankings requests carry the five weights as query parameters, so the backend scores with exactly what you see here.
Settings are stored in this browser only. They survive a reload and sync across tabs, but they are not uploaded anywhere and they are not shared with other users of the same backend.
Weights change how this model orders the universe. They do not change a single reported figure, they do not change what any company is worth, and they do not change any real-world outcome. A name that moves to the top of the table has not become a better investment; it has become a better fit for the preferences you just expressed.
What each category measures
Every metric below is computed from SEC XBRL filings and IBKR prices by deterministic arithmetic. No estimates, no forecasts, no model of anyone else's opinion.
Quality
Quality reads four figures straight out of the filings: operating margin, net margin, free-cash-flow margin and return on equity. A company scores well here when it turns revenue into operating profit, into net income and into cash, and when the equity base earns a high return. It says nothing at all about price, so a superb business can still be an expensive one.
- Operating Margin
- Net Margin
- FCF Marginfin. excl.
- Return on Equity
Needs Income-statement and cash-flow facts from the latest SEC filing.
Valuation
Valuation compares the current IBKR price against reported earnings, sales and free cash flow. Earnings yield and free-cash-flow yield score better when higher; price to earnings and price to sales score better when lower. Raising this weight tilts the ranking toward cheaper names, which is not the same thing as better businesses.
- Earnings Yield
- FCF Yieldfin. excl.
- P/E
- P/S
Needs A real IBKR price. With no price the whole category is N/A for that company, never zero.
Growth
Growth uses reported one-year changes plus three- and five-year compound rates for revenue, diluted EPS and free cash flow. Every figure is historical and comes from filings. Nothing here is a forecast, and no analyst estimate is used anywhere in the model, which is why forward P/E is permanently N/A.
- Revenue Growth YoY
- Revenue CAGR 3Y
- Revenue CAGR 5Y
- EPS Growth YoY
- EPS CAGR 3Y
- EPS CAGR 5Y
- FCF Growth YoYfin. excl.
- FCF CAGR 3Yfin. excl.
Needs Enough filing history to cover each comparison period. A short history leaves the longer CAGRs N/A.
Financial Strength
Financial strength weighs the current ratio and interest coverage, where higher is better, against debt to equity and net debt measured in years of free cash flow, where lower is better. Total debt is used only when short-term and long-term debt come from the same period end; when they do not, everything derived from it stays N/A instead of being stitched together across fiscal years.
- Current Ratiofin. excl.
- Interest Coverage
- Debt / Equityfin. excl.
- Net Debt / FCFfin. excl.
Needs Balance-sheet facts drawn from one consistent period end.
Risk
Risk scores Sharpe and Sortino better when higher, and annualized volatility, maximum drawdown, downside deviation and beta better when lower. It is computed from price history, so it describes the stock rather than the company. A name with no usable history is not punished for it: the category drops out and the surviving weights renormalize.
- Sharpe Ratio
- Sortino Ratio
- Annualized Volatility
- Max Drawdown
- Downside Deviation
- Beta
Needs Daily price history, plus benchmark history for beta.
With sector-aware scoring on, companies in the Financials sector are not judged on industrial leverage and free-cash-flow metrics: current ratio, debt to equity, net debt to free cash flow, FCF margin, FCF yield and both FCF growth metrics drop out, and the surviving weights renormalize.
Within a category, each present metric is turned into a cross-sectional percentile rank against the rest of the scored universe, ties averaged, mapped onto 0 to 100 and turned the right way up for its direction. A universe of one scores 50.0 on every metric it has, because a percentile against nobody is meaningless.
Portfolio and backtest options
Construction rules, selection filters and the cost assumptions applied at every fill.
Both commission and slippage are applied at execution, at the next trading day open after the signal. Neither is ever dropped to zero on your behalf.
Backtests remain price-return only and run on the current universe, so survivorship bias is present no matter what is configured here. Changing these options changes the model portfolio, not any holding you own.