SentalysisIn this guide
Key takeaways
- Market analysis rests on three pillars: technical analysis reads the chart, fundamental analysis reads the asset, and sentiment analysis reads the crowd.
- Sentiment is the pillar most traders handle by feel. It is not an unknown idea; it is an underdeveloped one, practised at four very different levels.
- Levels 1 to 3 (your gut, your feed, your poll) all sample people close to you, so they tend to echo what you already think.
- Level 4 measures the public at scale and compares every reading with its own history, so “unusual” becomes a number instead of a feeling.
- Sentiment measures intensity, not direction. A spike in attention says something matters to a lot of people, not which way the price goes next.
The three pillars of market analysis
Anyone trying to judge where a market might go is, knowingly or not, weighing three kinds of evidence. Each looks at the same price from a different side, and each has a blind spot that the other two help to cover.
TA Technical analysis
What is the price doing?
Price, volume, trends, support and resistance
FA Fundamental analysis
What should it be worth?
Supply, usage, adoption, interest rates, regulation
SA Sentiment analysis
How is the crowd behaving?
Attention, mood and discussion, and how unusual they are
Technical analysis (TA) reads the chart. It studies the history of price and volume (trends, ranges, support and resistance, momentum) on the idea that how a market has moved says something about how it is moving now. Its blind spot: a chart shows what people did, never why they did it.
Fundamental analysis (FA) reads the asset. For a company that means earnings and cash flow; for Bitcoin it means things like its fixed supply schedule, network usage, adoption, regulation and the interest-rate backdrop. It asks what something should be worth. Its blind spot: markets can ignore value for a very long time.
Sentiment analysis (SA) reads the crowd: how strongly people feel, what they are paying attention to, and whether that is ordinary or extreme. It covers the gap the other two leave. Prices are set by people, and people overshoot in both directions. A view that draws on all three pillars is harder to fool than one that leans on any single one.
Why sentiment is the overlooked pillar
Ask traders whether sentiment matters and nearly all of them will say yes. Ask how they measure it and the answers go vague: a feeling, a scroll through social media, a poll. Technical analysis has its indicators and fundamental analysis has its spreadsheets. For most people, sentiment has a mood.
That is not because sentiment is a foreign idea. Everyone has heard that you should be fearful when others are greedy. It is because there are levels to understanding it, and most people never get past the first two. The circles at the top of this page show the pattern: each level asks a wider group of people, and only the last one stops asking the people who already agree with you.
Level 1: the gut feeling
This is how most traders first meet sentiment. They know the contrarian sayings, “buy when there's blood in the streets” and Warren Buffett's “be fearful when others are greedy”, and they apply them by feel. A big red candle appears, anxiety follows, and the conclusion writes itself: this must be the blood.
The trouble is that the feeling is not independent evidence. It is the price, felt a second time. The red candle came first and the anxiety followed, so Level 1 can only tell you what the chart already said, filtered through one person's nerves. Its sample size is one, and that one person is the least neutral observer in the room: the one holding the position.
A gut feeling is not useless. Noticing your own fear is a fine reason to slow down. It just cannot tell you what anyone else feels.
Level 2: the social feed
At Level 2, a trader looks outward and asks what other people are saying. Scrolling social media, they pick up a bias from the mood of the posts: a mostly bullish timeline, or a wave of panic.
It feels like a survey of the market, but it is a sample chosen twice. First by you: your feed is made of the accounts you decided to follow, and they tend to share your views. Then by the algorithm, which ranks what you see by what it expects you to engage with. That is usually more of what you already agree with, delivered at its most emotional.
Level 2 is wider than Level 1, but it stays inside your own social circle and whatever the algorithm thinks you will resonate with. Two traders scrolling at the same moment can come away with opposite readings of the same market, and both will feel sure. Our guide to crypto social sentiment covers the other traps in discussion data: bots, sarcasm and coordinated campaigns.
Level 3: the poll
Level 3 is the first level that counts. Instead of soaking up the mood of a feed, a trader asks directly, with a poll on social media that invites followers to pick a side. That is a real step forward: a poll counts votes rather than impressions, and it gives quiet accounts, people who never post, a way to be heard.
But a poll inherits the limits of Level 2. The people who see it are the poll-maker's followers and whoever the algorithm shows it to, and the people who answer are the ones who chose to. So the result describes an audience, not a market. A perma-bull who runs a poll gets a very different answer from a perma-bear who asks the same question on the same day. Try it:
Their followers, the people who vote
69% lean bullish
Where does Bitcoin close the month?
Same question, same day, same market. The answer says more about a perma-bull's followers than about Bitcoin.
Illustration with invented numbers. Drag the slider to change who runs the poll. Only the audience changes, and the result follows it.
Polls are a good way to learn what your own audience thinks. As a reading of the market, they measure the person asking about as much as the market itself.
Level 4: the public, measured
Level 4 changes who decides the sample. Instead of your gut, your feed or your followers, the data comes from the general public, collected at scale, so the reading does not depend on who you follow, what the algorithm favours or who bothered to vote. And because a raw number means little on its own, each reading is compared with its own history: is attention to this topic ordinary, or the highest it has been in months?
This is the level Sentalysis works at. Its Attention Interest Oscillator measures public attention to tracked keywords across crypto, markets, geopolitics and macroeconomics, and the Z-score view shows how unusual each reading is. Here is Level 4 on two years of Bitcoin:
On 21 November 2025 and 6 February 2026, two of Bitcoin's major lows of the period, attention to Bitcoin spiked on the same day, to Z-scores of 4.6 and 5.3: four to five standard deviations above its recent normal. A gut feeling could not have put a number on that, and a feed or a poll would have depended on who you happened to follow. The turning-points study goes through every major high and low of those two years.
The four levels side by side
Every level answers the same question, how does the crowd feel, with a different crowd. The difference is who ends up in the sample, and who chose them.
| Level | You look at | Who is in the sample | Who picks the sample | Built-in bias |
|---|---|---|---|---|
| 1 · Gut | Your own reaction to price | One person: you | The last candle | Mirrors the price, so it adds nothing new |
| 2 · Feed | Posts on your timeline | Accounts you follow, plus what gets promoted | You, then the algorithm | Leans toward what you already think |
| 3 · Poll | Votes on one question | Followers who saw it and chose to vote | The poll-maker's audience | Leans toward the poll-maker's view |
| 4 · Measured | Public attention data | The general public, at scale | Nobody's follow list | Shows intensity, not direction |
How to move up a level
You do not have to give up the lower levels; you just have to know which one you are standing on.
Name the feeling, then check the chart
When a candle makes you anxious, say so, then ask whether the feeling holds anything the price did not. If it doesn't, it is Level 1.
Audit your feed
Follow a few sharp people you disagree with, and treat a one-sided timeline as a fact about your follows, not about the market.
Read polls as audience surveys
Before trusting a result, ask who runs the poll and who follows them. A lopsided answer from a lopsided audience is not news.
Measure, against history
Look at public attention to the topic and how unusual it is today. Sentalysis charts show Bitcoin on daily data for free.
Where Sentalysis fits
Sentalysis is built for Level 4. Sentiment charts put a crypto price beside the Attention Interest Oscillator for the keyword you choose, with every candle colored by how unusual attention is. The free Fear & Greed gauge reads the market's mood from attention to fear-related and greed-related topics. HiveMind takes Level 2 to scale, mapping discussion across social media into themes instead of showing you the slice your feed picks. Custom indexes and data alerts let you combine the keywords you care about and hear when they cross a level you set.
None of it tells you what to buy or sell. It gives the third pillar the same footing as the other two: data you can look at, compare and check.
FAQ
Questions and answers
What is sentiment analysis in trading?
Sentiment analysis (SA) is the study of how market participants feel and what they are paying attention to. It is one of the three pillars of market analysis, beside technical analysis, which reads the chart, and fundamental analysis, which reads the asset. It ranges from a trader's gut feeling to public data measured at scale.
What are the three pillars of market analysis?
Technical analysis studies price and volume history. Fundamental analysis studies what an asset should be worth: for a company its earnings, for Bitcoin factors such as supply, adoption, regulation and interest rates. Sentiment analysis studies the crowd: its mood, its attention, and how unusual both are.
What is the difference between sentiment analysis and technical analysis?
Technical analysis looks at what the market did: prices and volumes. Sentiment analysis looks at the people behind those prices: how much attention a topic is getting and how strongly people feel. Some price-based indicators are described as sentiment, but they still measure price; sentiment data is measured separately from it.
Are social media polls a good way to measure sentiment?
They are a step up from a gut feeling because they count votes, but the voters are the poll-maker's followers who chose to answer. A perma-bull and a perma-bear asking the same question on the same day will get different answers, so a poll mostly describes its audience.
What is the most reliable way to measure market sentiment?
Data collected from the general public at scale, so the sample does not depend on whom you follow or who votes, and read against its own history, so you can tell an ordinary reading from an extreme one. That is the fourth level of sentiment analysis.
Can sentiment analysis predict prices?
No method predicts prices reliably, and sentiment is no exception. It measures intensity and mood, not direction. Used alongside technical and fundamental analysis, it is context for your own research. Sentalysis provides research data, not trading signals or advice.