Stock Alert Confidence Scores: Five Questions Before You Act

A stock alert can give you something to investigate. It cannot remove the need to decide what the information means, what remains uncertain and whether it belongs in your research process.

Start with an important boundary: Foreshadow AI confidence scores are not calibrated probabilities of profit. Do not translate a score into an expected win rate or assume it tells you how much money an idea might make.

This five-question review helps separate a research input from a trading decision. You can use it without placing a trade.

1. What exactly am I trying to understand?

Write one specific research question before interpreting the alert.

“Should I buy?” bundles together several different issues: the evidence, your time horizon, the price available to you and your personal risk constraints. A narrower question is easier to examine: “What information would support or challenge my existing view of this stock?”

If you cannot state the question, record the alert as something to investigate rather than treating its arrival as a reason to act.

2. What time horizon am I evaluating?

Separate when information appeared, when you reviewed it and the period your research question concerns. Those are different things.

For example, an evening review of an earlier notification should not assume that the price or surrounding information is unchanged. Check the context available at your review time. Do not silently substitute today's circumstances for those that existed earlier.

Your notes should identify the relevant dates and times, including the time zone where needed. If the intended horizon is unclear, record that uncertainty instead of inventing one.

3. What evidence can I check independently?

Distinguish the information you have checked from an interpretation you have not verified.

For each important statement, ask what would let you examine it: a dated company announcement, a filing or market information appropriate to the question. Record what you actually consulted and when. If a key source is unavailable, mark the claim unresolved.

These are tasks for your own research process, not a promise that Foreshadow AI supplies every source, field or answer. A confidence score is not a substitute for checking the evidence behind a decision.

4. What would change my mind?

Write down a competing explanation or a condition that would weaken your interpretation.

Consider this hypothetical research note, not a Foreshadow alert or a reported result:

> Question: Does a recent company announcement change my longer-term view? > > Checked: I read the announcement and recorded its date. > > Unresolved: I have not compared the new information with the company's earlier expectations. > > Next step: Make that comparison before drawing a conclusion.

The useful output is not a prediction. It is a visible boundary between what has been checked and what still needs work. “Not enough information” is a valid research conclusion.

5. Am I keeping research separate from execution?

An alert, your interpretation and a broker order are separate steps. Foreshadow AI does not execute trades for subscribers.

Even an accurate observation does not, by itself, determine a suitable trade. Any eventual decision must account for your circumstances, potential losses and execution conditions. A confidence score does not settle those questions or establish that an idea will be profitable.

Use this compact note whenever you review an idea:

Question: What am I investigating?
Timing: When was the information available, and when did I review it?
Checked evidence: What did I verify?
Open questions: What remains uncertain or could challenge my view?
Next step: Research further, revisit later or take no action?

Use the seven-day trial to assess research fit

Foreshadow AI covers 520+ U.S. equities and delivers through Telegram and Discord. Its seven-day trial is an opportunity to assess whether the research fits your workflow—not to establish a dependable investment-performance expectation from a short period.

During the trial, try the note above on research you actually review. Can you explain the question, distinguish checked information from uncertainty and identify a sensible next step? You do not need to place trades to evaluate those aspects of usability.

For help choosing a review frequency, read Daily vs. Hourly Stock Alerts.

Compare Foreshadow AI plans and start a seven-day trial.

References

Foreshadow AI plans — plan comparison and trial destination.
Daily vs. Hourly Stock Alerts — a practical companion guide to research cadence.

Foreshadow AI provides market analytics and alerts for information and education. It does not execute trades for subscribers. Investing involves risk, including loss of capital. Past performance does not predict future results. Confidence scores are not calibrated probabilities of profit.

Explore Foreshadow plans and start your seven-day free trial