Performance Methodology

1. The three performance numbers

Three different measurements appear on this site, and two of them are routinely mistaken for the third. Every figure shown to a reader carries the label of the one it is. Where a measurement does not exist yet, the site prints "Not tracked" rather than estimating it or substituting another.

Weekly move. The stock’s price movement during the measurement window the issue analysed, from the official close on the last trading day before the window to the official close on the last trading day of it. That window closes BEFORE the issue publishes. Candidates are screened partly for having moved during it, so a strong weekly move describes the screen rather than a result, and it is never a return a reader could have earned by acting on the issue.

Publication price and publication date. The official close on the issue date, and that date. This is the reference basis for everything measured after publication, and it is the earliest price at which a reader could have acted.

Post-publication return. The price return from the publication price forward to the end of a stated window (currently one week and one month), on the same split-adjusted, dividend-excluded basis as every other figure here. A window is published only once it has completed.

Relative return. The post-publication return minus the benchmark’s return over the identical window, stated in percentage points rather than percent, because the difference between two rates is not itself a rate. The benchmark is the S&P 500 unless an article states otherwise.

Weekly Move (%) = ((Window End Close / Window Start Close) - 1) x 100

Post-Publication Return (%) = ((Close at Window End / Publication Price) - 1) x 100

Relative Return (pts) = Post-Publication Return (%) - Benchmark Return (%)

Historical issues are not retro-fitted. Where only a weekly move was ever measured for a pick, that is what is shown, labelled as a weekly move; no post-publication figure is reconstructed for it beyond what the verified price history supports.

2. Core principles

The Friday Five aims to present performance that is:

• Reproducible from identified data and timestamps.
• Clear about whether results are actual, hypothetical, paper-traded, backtested, or model-based.
• Inclusive of losing and closed examples, not only favorable outcomes.
• Adjusted consistently for corporate actions.
• Accompanied by assumptions, benchmark, costs, and limitations when relevant.

3. Security identity and eligibility

Before publication, verify:

• Company or fund name.
• Ticker and primary exchange.
• Security type.
• Measurement dates and market holidays.
• Corporate actions, symbol changes, mergers, spin-offs, splits, special dividends, and delistings.
• Liquidity and whether the displayed price was reasonably executable.

Unless stated otherwise, rankings cover common stocks and exchange-traded funds listed on major U.S. exchanges. Excluded categories should be identified, such as over-the-counter securities, warrants, rights, preferred shares, leveraged products, or securities below a stated price or liquidity threshold.

4. Weekly performance

Unless an article states otherwise, weekly price performance is calculated from the official closing price on the final trading day of the prior week to the official closing price on the final trading day of the measured week:

Weekly Return (%) = ((Ending Price / Starting Price) - 1) × 100

If a market holiday changes the trading week, use the last available official close in each period and state the dates.

5. Monthly performance

Unless stated otherwise, monthly price performance is calculated from the official close on the last trading day before the measured calendar month to the official close on the final trading day of the measured calendar month:

Monthly Return (%) = ((Month-End Price / Pre-Month Price) - 1) × 100

Monthly tables list every stock covered in that calendar month, not only the leaders. A "Top Five" table may appear as a highlight, but it must sit alongside the complete list for the same month, never in place of it.

5a. Since-inception figures

The since-inception summary on the Scoreboard measures each published pick over the issue week it appeared in, and compares it with the S&P 500 over that identical window. Figures are calculated as follows:

• Picks published: every pick slot across every issue. A company selected in two different issues counts twice, because each selection was a separate call.
• Average pick week: the simple, equal-weighted mean of those issue-week returns. It is not compounded, not capital-weighted, and not a portfolio return.
• Benchmark: the equal-weighted mean of the S&P 500 price return over the same set of issue weeks, so an issue week appears in the benchmark once per pick published that week.
• Win rate: the share of picks with a positive issue-week return. Beat rate: the share that exceeded the S&P 500 over that same week.
• Excluding buyouts: the same averages with picks removed whose move came from an announced bid for the company itself.

An average of single-week returns drawn from different weeks is not a return an investor could have earned. It ignores position sizing, costs, taxes, and the order in which weeks occurred. It describes what was published. It is not performance, and a history this short cannot establish skill.

Monthly tables are ranked by price return alone. A pick whose move came from an announced acquisition of the company is labeled as such, because that return is a single unrepeatable event that would otherwise sit at the top of a ranking permanently and overstate the record.

6. Price return versus total return

Every presentation must identify whether it uses:

• Price return: change in market price, adjusted for splits but excluding cash distributions; or
• Total return: change in value including reinvested distributions according to the identified methodology.

The default Friday Five stock-ranking metric is price return using split-adjusted closing prices, excluding dividends, unless the article labels a different method.

Do not mix price-return and total-return figures in the same ranking without clear labels.

7. Corporate actions

Stock splits and reverse splits should be normalized so the action alone does not create a false gain or loss. Special dividends, spin-offs, mergers, and rights distributions require case-specific review and a methodology note.

A security may be excluded when a corporate action prevents a fair comparison.

8. Intraday references

Intraday returns, entries, targets, and stops must state:

• Date and time.
• Time zone.
• Data source.
• Whether the price is bid, ask, last trade, midpoint, volume-weighted average, open, high, low, or close.
• Whether the level was published before or after it occurred.

A price observed after publication may not be represented as an available entry unless the assumptions explain why execution was feasible.

9. Model trades and portfolios

A model trade is not a subscriber account and does not represent individualized advice. Each model must state:

• Publication timestamp.
• Security and instrument.
• Entry rule and assumed price.
• Position-sizing rule.
• Exit, target, stop, or expiration rule.
• Benchmark.
• Fees, spread, slippage, taxes, and dividend assumptions.
• Whether partial shares are assumed.
• Whether capital is continuously available.

If assumptions are not included, the presentation must not be marketed as a complete track record.

10. Options methodology

Options presentations must identify:

• Underlying symbol.
• Strategy and position legs.
• Expiration date.
• Strike prices.
• Option type.
• Assumed contracts and multiplier.
• Premium basis: bid, ask, midpoint, last, or executable fill.
• Commissions, contract fees, spread, slippage, assignment, exercise, and early-exercise assumptions.
• Maximum defined gain and loss when calculable.
• Treatment at expiration and of early exits.

A multi-leg strategy may not be marked using favorable prices from different moments unless the methodology expressly models simultaneous execution and realistic spreads.

11. Actual, hypothetical, paper, and backtested labels

Use one of these labels prominently:

• Actual publisher account: supported by internal records; not independently audited unless expressly stated.
• Paper trade: recorded prospectively without real capital.
• Hypothetical model: calculated from stated rules without actual execution.
• Backtest: rules applied retrospectively to historical data.
• Illustrative example: educational arithmetic, not a track record.

Never present hypothetical, paper, backtested, or illustrative results as actual trading results.

12. Costs and taxes

Unless expressly included, headline stock-ranking returns exclude commissions, spreads, slippage, market impact, borrowing costs, subscription fees, and taxes. Any portfolio or strategy performance should state which costs are included.

Tax treatment varies by investor and may materially change after-tax results.

13. Benchmarks

When comparing performance, identify the benchmark, ticker or index, return type, dates, and whether dividends are included. The benchmark period and return method should match the presented strategy as closely as practical.

14. Verification process

Before publication, a review step separate from the one that selected and wrote up the security should confirm:

1. Ticker and security identity.
2. Start and end dates.
3. Start and end values.
4. Return formula.
5. Corporate-action treatment.
6. Liquidity and data anomalies.
7. Catalyst attribution and source dates.
8. Labels for actual, hypothetical, paper, backtested, or illustrative results.
9. Conflict and position disclosures.
10. Consistency between article text, charts, tables, PDF, email, and website.

What "verified" means here, and what it does not. Verification means a figure was reconciled against more than one market data source and traced to a primary or reliable secondary source, by a step in our process separate from the one that produced it. It is separation of duties inside one publisher.

It is not an independent audit. No outside firm, auditor, exchange, or regulator reviews these figures, and none of them attests to this record. Where The Friday Five describes a number as checked or verified, it should be read in that narrower sense. Claims of independent or third-party verification must not be made unless an outside party has actually performed the review and is named.

15. Corrections

Material calculation or classification errors will be corrected under the Corrections Policy. The correction should identify the original value, corrected value, reason, and date.

16. Limitations

Historical results do not guarantee future outcomes. Rankings are retrospective and can be affected by survivorship bias, selection rules, data revisions, illiquidity, and events that could not have been known in advance.