Best Alphanume Research Posts on Options and Volatility

Options research usually arrives in the wrong order. A trader starts with a payoff diagram, finds an attractive backtest, and only later asks whether the quote, volatility measure, or distribution was observable when the trade supposedly happened.

This three-post Alphanume Research path reverses that sequence. It starts with a dated probability input, moves to portfolio accounting, and finishes with the plumbing that determines the fill.

We reviewed the public Alphanume Research archive on September 4, 2026. Every selection is affiliated with Alphanume because this is a guide to Alphanume's own work. We selected posts for distinct research jobs, methodological detail, and usefulness beyond the publication date. We did not rank them by returns or claim that their examples will persist.

How the three posts were selected

A useful options library should answer three questions in order:

1. What did the option market imply at the decision time?
2. How did the strategy or product transform that exposure into investor returns?
3. What price and route could a real order obtain?

The posts below map one-to-one to those questions. They also expose enough process to build a reproduction plan. A memorable return number was not an inclusion criterion.

First: build the signal from information that existed

SPX 0-DTE Options Are Still a Cash Cow centers on the strike-selection problem for same-day SPX options. Its useful research object is not the phrase cash cow. It is a lower and upper strike published at a specified time from a model using information available then.

That setup creates testable questions. How often did the closing price remain inside the band? How often did the intraday high or low cross it? Did results differ on CPI, FOMC, expiration, or high-volatility days? How much premium was actually available after the level was published?

Pull the SPX 0DTE Strike Band and save the retrieval timestamp. Join each band to quotes strictly after publication. Report band width, close containment, intraday containment, option spread, and worst loss. A close-only success rate cannot describe the path risk of a short option.

Second: reconcile income with total return

0-DTE Income ETFs Are a Disaster... Sort Of asks a better question than whether an options ETF has a large distribution yield. It examines what the investor kept after distributions, changes in net asset value, and market exposure.

The accounting identity is simple:

ending_wealth = ending_price + cash_distributions
total_return = ending_wealth / starting_price - 1

# For reinvested comparisons, use adjusted values or model each payment.
# Keep tax treatment separate because investor circumstances differ.

The research work is in cohort construction. Separate daily or weekly overwrite funds, index versus single-name exposure, capped-upside structures, and funds with different distribution policies. Use prospectus rules as features. Compare each product with a benchmark that matches its underlying beta and option exposure rather than a generic equity index.

This reading is most useful for learning how a product's headline yield can diverge from economic return. It does not establish that every covered-call or 0DTE fund is good or bad.

Third: understand where the fill comes from

A Cracked Quant's Guide to Orderflow follows a retail stock order through broker routing, wholesalers, exchanges, and public disclosure. The instrument in the opening example is stock, yet the lesson matters even more for options because spreads, multi-leg execution, and fragmented liquidity can dominate a small theoretical edge.

A backtest midprice is a hypothesis about execution. For a one-leg strategy, compare midpoint, bid or ask, and a conservative fraction of the spread. For a multi-leg strategy, test package execution and legged execution separately. Add commissions, exchange fees, and rejected or partial orders.

Order-flow analysis also needs direction humility. A printed option trade does not automatically reveal the initiator's full position or whether another leg exists. Use public tape fields as evidence, not a complete portfolio reconstruction.

The datasets that connect the readings

The three posts become more useful when paired with research-ready features. The IV Rank dataset places current implied volatility inside its own history. The IV versus historical-volatility premium dataset compares implied and realized measures. Neither variable is a trade by itself.

A clean combined study might ask whether wide SPX strike bands, high IV rank, and a large IV-to-HV premium identify different regimes for same-day premium selling. Pre-declare the thresholds, use the version of each value available before entry, and retain days when one field is missing.

Avoid stacking filters until the backtest looks perfect. Every threshold, expiry, entry time, and exclusion increases the search count. Reserve a later period before tuning and report how many variants were attempted.

A compact robustness grid can cross three entry times, three spread assumptions, and two volatility definitions. Mark the original specification before running the grid. The goal is not to find the best cell. It is to see whether the conclusion survives reasonable choices around the original one.

Use a reproduction card for each post

Complete the card before downloading the final outcome series. The exercise exposes vague rules while changes are still cheap.

A four-session reading and testing plan

1. Read the strike-band post and write the publication-time join rule.
2. Pull a small date sample and verify every band against the first eligible quote.
3. Read the ETF post and reconcile price return, distributions, and total return for one fund by hand.
4. Read the order-flow guide and replace midpoint fills with two conservative execution models.

Only then scale the sample. Manual checks on a few dates catch timezone errors, stale quotes, adjusted-price mistakes, and distribution double counting before they contaminate years of results.

Archive the input rows used in each manual check. A one-page audit packet should show the source timestamp, eligible quote, selected contract, corporate-action adjustment, cash flow, and calculated return. Those packets become regression tests when the pipeline changes.

What to read for which problem

Start with the SPX post if your question is signal timing or strike selection. Start with the ETF post if you are comparing income products or trying to understand distribution yield. Start with the order-flow guide if a paper edge disappears in live trading.

The sequence matters for a complete project: signal, portfolio accounting, execution. If any layer is missing, the conclusion is provisional.

The Alphanume Research archive contains the broader catalog. Use it to find a mechanism, then return to the datasets and write a falsifiable test. A good reading list should end in a research specification, not a larger bookmark folder.

All three selections are Alphanume publications. The examples are educational research and not personalized investment advice.