Collateral, premium, period and annualized return, breakeven, and cost basis if assigned for a cash-secured put, including what the parked cash can earn in T-bills or a money-market fund.
Annualized return on collateral
n/a
Collateral requiredn/a
Premium collected (net of commissions)n/a
Return on collateral for the periodn/a
Annualized premium returnn/a
Total annualized incl. T-bill yieldn/a
Breakeven at expirationn/a
Downside cushionDistance from the stock price to breakevenn/a
Simple (non-compounded) annualization on a 365-day year. Assumes the put is held to expiration; ignores early assignment, assignment fees, and taxes.
Dividends
Ex-Dividend Drop Estimator
Estimates the ex-day opening price using our Ex-Dividend Day Price Drop study: across 5,106 ex-dividend events in 80 S&P 500 stocks (2010–2025), shares opened lower by an average of 0.812 times the dividend (95% CI 0.762 to 0.861) before adjusting for market moves. Once the market's own move is correctly backed out, the paper finds the average drop is roughly the full dividend, so these raw ratios partly reflect the market's direction on those days.
Expected drop at the open
n/a
Drop ratio usedn/a
Expected ex-day open pricen/a
Expected open, rangen/a
Theoretical full-drop pricePrior close minus the full dividendn/a
Shortfall vs. a full dropn/a
Formulas
Expected drop = dividend × drop ratio
Expected open = prior close − expected drop
Full-drop price = prior close − dividend
Yield = dividend × payments per year ÷ price
Drop ratio (paper) = (prior close − ex-day open) ÷ dividend
Historical average; individual events vary widely. The drop ratio's standard deviation across events was 1.83, so a single stock's open can land far from this estimate. The range reflects uncertainty in the average ratio, not the spread of individual events. Adjusted for market moves, the paper's average rises to about 1.0, the full dividend. Read the paper.
Technical analysis research
RSI Study Explorer
Interactive results from our RSI study: four Wilder RSI(14) strategies backtested across 100 U.S. stocks in 11 sectors with up to 64 years of daily data (1962–2026), 5,174 trades in total. Buy when RSI crosses below the lower threshold, sell when it crosses above the upper threshold.
Read with care. The 100-stock universe consists only of companies that survived to 2026 (delisted, bankrupt, and acquired firms are excluded), which biases win rates and average returns upward; the paper does not estimate the size of this bias, which is likely largest for long-hold strategies such as Weekly 20/80. The tests cannot separate RSI timing from the upward drift of these surviving stocks. Stock by stock, every strategy lagged buy-and-hold on an annualized basis (paired t-test p < 10⁻¹⁹ for all four), mainly because capital sat idle between signals; that comparison uses single stocks with no return on idle cash, not a portfolio or an index. These are hypothetical backtest results, not investment advice.
Strategy results
Table 1 and Table 2 of the paper. Toggle strategies and pick a metric.
By sector
Tables 3 and 4. The paper publishes sector tables for the two 30/70 strategies only, and the rankings reflect today's surviving winners; for the 20/80 strategies, sector differences were not statistically significant (ANOVA p = 0.8226 daily, 0.5812 weekly).
Show sector table
Market regimes
Table 6: performance during market crises, filtered by the strategy toggles above. Some cells rest on very few trades (Weekly 20/80 had 2 trades in the 2022 bear market).
Run the paper's exact entry and exit rules on any daily price history. Upload or paste a CSV with a Date column and a Close or Adj Close column, such as a Yahoo Finance historical-data download. The file is read in your browser and never uploaded.
First row must be headers. Dates as YYYY-MM-DD or MM/DD/YYYY; any row order.
Trades
n/a
Win rate
n/a
Avg return / trade
n/a
Avg holding (days)
n/a
Strategy return (compounded)
n/a
Buy-and-hold return
n/a
Time in market
n/a
Span (years)
n/a
Rules (matching the paper's methodology)
RSI uses Wilder's smoothing: average gain and loss are seeded with a simple average of the first N changes, then updated as (previous × (N − 1) + current) ÷ N. The paper's implementation (an exponential average with α = 1/N) converges to the same values after a short warm-up, so the two can disagree only on signals near the start of a series.
Entry: buy at the close when RSI is below the lower threshold and the prior bar's RSI was at or above it (a downward cross). Exit: sell at the close when RSI is above the upper threshold and the prior bar's RSI was at or below it (an upward cross). One position at a time; a position still open at the end is closed at the last price and flagged.
Returns are gross (no commissions, slippage, or taxes). Buy-and-hold runs from the first to the last close of the same series; time in market is calendar days held divided by calendar days in the span.
Financial modeling
DCF Modeler Now free
A per-share discounted cash flow model with Bear, Base, and Bull scenarios, share buyback modeling, a sensitivity grid, and a Monte Carlo simulation, all in one HTML file.
Start from a company's free cash flow (total or per share) or its EPS. The modeler projects that figure forward with a year-one growth rate that decays each year, shrinks the share count by your buyback rate, discounts each year's per-share value at your discount rate (WACC), and adds a terminal value from an exit multiple on the final year.
Bear and Bull cases are offset from the Base case by adjustable growth, multiple, WACC, and buyback deltas, so you see the range at a glance and how each sits against today's price. It is a single self-contained HTML file with no external scripts: open it in any browser, online or off.
Metric basis: total FCF, FCF per share, or EPS, valued per share
Bear, Base, and Bull scenarios with adjustable offsets
Year-one growth with annual growth decay over a 3 to 20 year projection
Discount rate (WACC) and exit-multiple terminal value
Share buyback modeling: an annual buyback rate lowers the share count and lifts per-share value
Sensitivity table: a 5 × 5 grid of growth rate vs. exit multiple, shaded against the current price
Monte Carlo simulation: 2,000 runs randomizing growth, WACC, exit multiple, and buyback, with a histogram and 5th to 95th percentile values
Year-by-year projection table and a copyable step-by-step valuation equation
CSV export of all three scenarios and the base-case projections