Signal Methodology and Risk Disclosure

1. Shared publication and regular schedule

WealthGlider's standard publication schedule is one shared edition per scheduled U.S. market day after the scheduled model run. The edition is created for the eligible audience as a whole, not in response to a particular user's request, portfolio, or circumstances. Market holidays, unavailable or defective source data, security incidents, and maintenance can delay or prevent an edition; we do not create ad hoc editions or bespoke recommendations to fill a missed cycle.

Each edition has one underlying set of published records. Access level may limit how much of that edition or the archive a user can view, but it does not change a published security's label, score, explanation, or history for that user. Published editions are maintained by date, and a correction should preserve the original publication record, correction time, and reason rather than silently rewriting history.

2. Impersonal service and API delivery

WealthGlider publishes model-generated securities research rather than tailoring it to an individual's circumstances. We do not ask for or use your portfolio, holdings, prior trades, brokerage credentials, investment objectives, income, net worth, tax situation, time horizon, or risk tolerance to generate or alter a signal. Authentication, entitlement, security, and rate-limit information control access to the service; they do not personalize model output.

Selecting a ticker or date retrieves the same existing published record available to any other eligible user making the same request. The API is another delivery method for the shared publication and archive; it does not accept portfolio or suitability inputs and must not generate an individualized opinion. WealthGlider does not offer real-time advisory chat, bespoke ticker opinions, or individualized event-triggered signal timing.

We do not execute, route, or recommend a broker for a trade, hold customer assets, monitor a portfolio, decide whether an individual should trade, or agree to act as a fiduciary. If you need a recommendation suitable for your circumstances, consult a properly licensed professional who can evaluate your complete situation.

3. What Buy and Sell mean

"Buy" and "Sell" are names for research classifications produced by WealthGlider's model workflow. They describe the direction of a model-selected candidate under a defined 30-trading-session evaluation method. They are not an instruction to place an order, a statement that a security is suitable for you, or a promise that the security will move in that direction.

The displayed ranking score is an internal composite used to order model candidates. It is not a probability that the target will be reached and is not a measure of your expected return. A separately labeled historical target-hit rate describes completed historical records only.

4. Reasonable basis for factual product descriptions

We limit public product descriptions to functions that can be inspected in the service:

• The dashboard displays model-generated ticker classifications, a research category, a score, and model-driver information when supplied by the data pipeline.
• The dashboard, date tools, symbol search, and API retrieve records from the shared publication and archive; they do not accept a user's financial profile as a model input.
• “Built to learn the market” means the model uses statistical training on market data to identify patterns. It does not learn from an individual user's identity, portfolio, searches, or trades, and statistical training does not guarantee accurate or improved future results.
• Each record uses the official closing price on its publication market date as its reference price. Its evaluation price is exactly 10% above that close for Buy or 10% below it for Sell.
• The evaluation window is the next 30 observed trading sessions; the publication session is not counted. A threshold is evaluated against the observed intraday high for Buy or intraday low for Sell. Later prices are adjusted for stock splits into publication-date share units. Cash dividends are not included.
• The Historical Signal Evaluation is updated each market day when eligible price-path data is available. Publication history and evaluation coverage are different: recommendation archives include editions from August 7, 2026, while each evaluation identifies its supported history boundary and data-update date. A record becomes Hit when the applicable threshold is first touched within the 30-session window, even before the window ends. With no target touch, it remains TBD until all 30 sessions are observed, then becomes Miss. Invalid or insufficient price data is unscorable. An observed resolved-record target-hit rate is Hits divided by Hits plus Misses in the selected group and period, excluding TBD and unscorable records. The backtesting chart displays a provisional score of 100 × (hits + 0.5 × pending) / (hits + misses + pending). Existing TBD records receive half credit, and unscorable records remain excluded. Half credit is a scoring convention, not an estimated probability of success; the Hit/Miss/TBD classifications do not change.
• Early Hits resolve before pending records can become Misses. An immature cohort's target-hit rate can therefore overstate its eventual rate. Fully mature comparisons require a complete 30-session window for every included publication, including early Hits, and separate missing-data counts.
• A target touch does not tell you whether the opposite 10% threshold was reached first or whether the price finished on the predicted side after session 30. Those are separate research outcomes. See the Results definitions and fictional worked example.
• SHAP values, when displayed, describe how model inputs contributed to a particular model output. They do not establish causation, correctness, or future performance.

These mechanics support statements about what the software displays and calculates. They do not support claims that WealthGlider is accurate, unbiased, profitable, superior to another strategy, or suitable for a user.

5. Historical and backtested information

Historical target-hit data and backtests are hypothetical or observational analyses, not results achieved in your brokerage account or in a portfolio managed for you. They may be revised when data, eligibility rules, model versions, or error corrections change. Unless a display expressly says otherwise, results do not include brokerage commissions, bid-ask spreads, market impact, slippage, borrowing costs, taxes, subscription fees, liquidity constraints, rejected orders, or the effect of position sizing and compounding.

Backtests can benefit from hindsight and may contain survivorship, selection, look-ahead, data-snooping, or overfitting effects. A target being touched during a historical window does not establish that a user could have entered or exited at that price; the calculation is a price-path observation, not an execution claim. Past or hypothetical results do not predict future results.

6. Data, model, and delivery limitations

• Market and reference data may be delayed, incomplete, adjusted, corrected, unavailable, or inconsistent across providers.
• Models may be wrong, unstable, overfit, affected by market-regime changes, or unable to account for news, halts, liquidity, corporate actions, or exceptional events.
• Scores and explanations simplify a complex model and should not be interpreted as a complete statement of risk.
• Web, API, or agent delivery may be delayed or fail. A scheduled edition may be late, corrected, or omitted when source data or systems are unavailable.
• Short sales, options, leveraged products, and thinly traded securities can create losses greater than an initial expectation and may not be appropriate for many people.

7. Your responsibility

Verify information using independent sources, review issuer filings and current market conditions, consider liquidity and total trading costs, and decide whether to act only after evaluating your own circumstances. Never trade money you cannot afford to lose. WealthGlider does not guarantee accuracy, completeness, timeliness, a target event, profit, loss avoidance, or uninterrupted availability.

8. Conflicts, issuer compensation, and personnel trading

WealthGlider does not accept issuer compensation to include, exclude, rank, or assign a direction to a security. A brokerage, data provider, advertiser, affiliate partner, or other third party cannot purchase a model label, ranking, explanation, publication time, or favorable discussion. Advertising and affiliate compensation do not determine model output. Sponsored and affiliate relationships are identified separately under our Affiliate Disclosure and Editorial Policy.

Personnel and contributors with prepublication access are subject to written holdings-disclosure and restricted-trading controls. They must disclose material interests in securities discussed in authored content and may not front-run, scalp, or trade using a nonpublic edition, model change, correction, or release schedule. A covered security is restricted from the time a person receives material nonpublic release information through the end of the next full U.S. market day after publication, subject only to documented compliance exceptions for non-discretionary transactions.

Report suspected issuer influence, undisclosed holdings, trading-policy violations, or inaccurate disclosures to support@wealthglider.com. Reports are reviewed and material corrections are documented.

9. Related terms

This disclosure is part of the Terms and Conditions. The service is directed to adults in the United States. Access from another jurisdiction does not mean the service is lawful, suitable, or offered for sale there.