What an AI stock score is
A stock score turns many facts about a company into one number, usually from 0 to 100. The facts can be price moves, risk measures, profit margins or valuation ratios. A formula combines them. Higher means the stock looks better on those inputs.
The "AI" part varies by product. In some tools a model writes the score. In others, a fixed formula does the math and AI explains the result. Ask which one you are looking at, because the answer decides how repeatable the number is.
The six parts of the Peakwise score
The Peakwise AI stock score compares each stock with a fixed list of 50 large S&P 500 companies. Each of six parts is a rank within that list, from 0 to 100. The score is the weighted average. Full details sit on the score methodology page.
- Momentum, 30%: price change over 3, 6 and 12 months
- Trend, 20%: price against its 200 day average
- Volatility, 15%: how calm the price moves were, calmer ranks higher
- Drawdown, 15%: how deep the worst fall was, smaller ranks higher
- Quality, 10%: net profit margin and debt relative to equity
- Valuation, 10%: earnings yield and free cash flow yield
Why a fixed comparison list matters
A score is relative. A 72 doesn't mean "72% good". It means the stock ranked well against the other stocks on the list. Change the list and the score changes.
That is why Peakwise uses a fixed list and says so. The list is not the whole market, and a stock outside it isn't scored. The upside is that two scores from the same day are comparable with each other.
Reading a score (illustrative)
Say stock A has a score of 78 and stock B a score of 52. A ranked better than B on the six parts. That's all you know. You don't know why yet. Maybe A had a strong year and B a rough one. Open the parts and look.
What a score can tell you
A score is good at triage. With 50 stocks on a list, sorting by score shows where to look first. It also shows which parts drive the number. A stock with strong momentum and weak valuation reads differently from one with the reverse.
You can also compare a stock with its own past, if the tool keeps history. Falling scores can flag that something changed.
What a score cannot tell you
It cannot tell you where the price goes next. Momentum and trend look backward by design. A high score describes how the stock behaved. It doesn't promise a repeat.
It cannot know your goals, your time frame or your other holdings. And it knows nothing the data doesn't contain: a lawsuit, a product recall or a new competitor may not show up until prices move.
It is also not an analyst rating. Analyst ratings are opinions published by research firms. The Peakwise score uses no analyst opinions and no target prices.
How to use a score in your research
Treat the score as the first filter, not the last word. Sort the list, pick a few names, then open a full analysis for each. Check the reasons behind the number.
Peakwise shows the score with its parts in the AI stock finder, and you can browse the ranked list on the stocks page. Data is end of day, so the page shows the date it applies to.
- Sort by score to build a short list
- Open the parts to see what drives each number
- Run an analysis on the names that remain
- Decide on your own, with your own time frame
Frequently asked questions
What is a good AI stock score?
A higher score means the stock ranks better than the others on the list. In Peakwise, 65 to 79 carries the label Strong profile and 80 to 100 Very strong profile. The labels describe the range. They are not recommendations.
Is an AI stock score a buy signal?
No. A score summarizes past prices and a few ratios. It doesn't know your goals and doesn't predict returns.
How is an AI stock score different from an analyst rating?
An analyst rating is an opinion from a research firm. The Peakwise score comes from a fixed formula on end of day prices and ratios, with no analyst input.
How often is the Peakwise score updated?
Once per trading day, after the market has closed, from end of day prices. The date is shown above the table.
Which stocks does the score cover?
50 large S&P 500 companies on a fixed list. It is not a screen of the whole market, and every score is relative to that list.