Chapter 4 · Why trends form
Why does a price rise often continue for a while?
- Skills to practice
- Understand marketsResearch Edge
- 3D simulation
- None
Market scene
On Monday morning, a major payments company announces that users will be able to pay in BTC. Analysts estimate that this is worth a 5% increase in BTC's price.
That day, BTC rises 3.75%.
The next day, social media fills with screenshots of the rally. Many people who never read the announcement see the rising price and buy. BTC reaches 8.81% above its pre-announcement level.
Day three: 13.55%. Day four: 16.9%. Day five: 18.32%.
Then the rise stops. Day six returns to 17.7%; day eleven to 10.69%; day twenty-one to 6.12%. After a month, BTC settles 5.26% above the original price, close to the analysts' initial estimate.
Why did news worth 5% produce five rising days and an 18% gain? If price eventually returned near 5%, who drove the intervening rise and decline?
Your decision
On the second evening, BTC is already up 8.81% from before the announcement. What do you do?
Observe the result
- Price rises far beyond what the news is worth. The news is worth 5%, yet price reaches 18%. The extra rise comes from people buying because price has risen.
- It eventually returns to the news-implied value. Almost all the extra gain disappears within a month.
- Entry timing matters more than simply deciding to buy. People sharing the same view have very different outcomes entering on day two or four. A later buyer increasingly provides an exit for earlier buyers.
- The choice with an exit rule has the steadiest result. It earns less than the maximum, but does not require knowing the peak.
Be careful with that last observation: you can now see the whole month's path. On the decision evening, you cannot. You do not know whether the move will last five days or five weeks.
The mechanism
Three forces drive the price:
- Buying prompted by the news. Strongest on day one, then progressively weaker.
- People buying because price has risen. The more it rose yesterday, the more they buy today. They watch price rather than news.
- People with a valuation. They value the news at 5%. The further price rises above that level, the more they sell; below it, they buy.
In the early days the first two forces combine to push price upward. The second creates a self-reinforcing loop:
- Price rises
- More people notice
- They follow and buy
- Price rises further
↺ Price rises · Rising prices attract buyers
The loop cannot continue forever:
- News-driven buying diminishes each day.
- The further price moves above 5%, the more the third group sells.
- Once a day fails to rise, followers stop buying and the loop breaks.
Day five is the turning point here: followers' purchases can no longer absorb valuation-driven selling. Price then falls toward 5%, where the three forces balance again.
Another limit is how much money the followers still have. In this model they can invest $2 billion. If only $500 million remains, they exhaust it on day three and the gain peaks at 10.62%. How far a rally travels depends on how many people have yet to enter.
What it is called
Assets that rose tend to keep rising for a time; assets that fell tend to keep falling. Momentum comes from price followers and from gradual digestion of information.
Money entering or leaving a market over an interval, on a net basis. Flows move price: the same news produces a larger move when it brings more buying for longer.
How many participants already hold an asset, who they are, and how they entered. The $2 billion here is money not yet invested. More people already aboard means less remaining money to drive the next move.
The story people use to explain price changes. A compelling, widely shared story attracts more enthusiastic buying. A narrative is not value itself, but changes flows.
An outcome strengthens its cause: rising prices attract buying, which raises prices further. This carries trends beyond fundamentals and can make reversal abrupt when fuel runs out. The liquidation cascade in Chapter 18 is downward positive feedback.
Real markets
GameStop traded below $20 in early January and reached an intraday $483 before month-end. Retail traders on online forums rallied around a story of making short-selling institutions lose money. At one point, shorted shares exceeded the stock's outstanding float.
The narrative generated buying, which forced shorts to cover—another purchase. Weeks later the fuel ran out, and the price gave back well over half its gain.
Spot Bitcoin ETFs were approved on January 10. During the preceding months, many people bought in anticipation. After approval, BTC fell about 20% over two weeks.
Most people wanting to buy for that news had already bought. When it finally arrived, little new money remained to push price.
Academic research across dozens of futures markets—equity indexes, bonds, commodities, and currencies—has found that markets rising over the previous year tend to rise over the next month, and falling markets tend to keep falling.
Many trend-following funds use this phenomenon. It is an average tendency, not a guarantee on every occasion: they also lose when trends reverse.
Hands-on
This is the chapter's market. Default parameters reproduce the month in the opening scene.
Blue shows the gain from the pre-news price. The gray dashed line is the news fundamental value of 5%. The orange vertical line marks when followers exhaust their capital.
On day 1, the news brings USD 300 million of buying, halving each subsequent day. Every USD 100 million of net buying moves price 1%. This simplified model illustrates the shape of a trend.
Course versionV1-docs; sourcelab:trend；Chapter 4 / TRD-MKT-004
Records parameters and results at the click only; does not mean the experiment passed.View snapshot to save
- Leave parameters unchanged. Record the maximum gain, its day, and the level after 30 days.
- Change only followers' available money from $2 billion to $500 million, then $4 billion. How does the peak change? Why does $4 billion make almost no further difference?
- Set the strength of valuation-driven traders to 0. Does price come back? Why?
- Find a rally triggered by news. Record the news and timestamp, then the next ten trading days' closes on a chart. How many consecutive days rose? How far did price fall from its peak?
Change one variable
They spend it all by day three. The maximum gain is only 10.62%, nearly half the original, before returning near 5%.
Trend length often depends less on how good the news is than on how many people have yet to enter.
The gain reaches 23.70% on day four, exhausts all followers' money, and falls back to 5.28% after a month.
A stronger narrative produces a higher surge and burns fuel faster. The eventual destination is still the news-implied value.
Nobody sells because price is expensive. Followers buy until all $2 billion is spent. The gain stops at 26% and never returns.
Valuation-driven participants pulled price back to 5%. A market containing only followers and nobody assessing value can remain far from value.
Three depths
- FoundationWhy does a price rise often continue for a while?Chapter 4
- AdvancedHow do you research Time-series Momentum: define signals and explain returns?Advanced B · Strategy research
- InstitutionalWhat is trend-strategy capacity, and how does crowding erode returns?Institutional
Distinguish the three forces: is the rally driven by the news itself or by followers? How many people have yet to enter?
Trend trading captures the middle: specify an exit before entry. Do not try to buy the bottom and sell the top.
Time-series Momentum: turn “hold after a rise over the past N days; stay out or short after a decline” into a signal. Size by volatility and test across markets and periods. Identify whether returns come from gradual information digestion or feedback from follower flows, and which market states break the strategy.
See Advanced B · Strategy research.
Continue the artifact: Freeze a time-series momentum signal and reversal test.
Crowding and capacity: when many trend funds hold the same position, reversal makes them exit together. The decline can outpace the rise. Risk then comes substantially from others being on your side.
The portfolio role: trend strategies often profit in crises—a sustained decline is also a trend—and slowly lose in choppy markets. Institutions combine them with other return sources and allocate capital through risk budgets.
Continue the artifact: Submit scale limits and constraints on crowded exits.
Questions to take away
Chapter self-test
News-driven buying is joined by price followers. Their buying raises price, which attracts more buying and pushes it beyond the news-implied value.
News-driven buying fades, and valuation-driven selling grows as price moves away from 5%. When price stops rising, followers stop buying. Valuation-driven participants then pull price back near 5%, restoring balance.
It is followers' money that has not yet entered. Less fuel ends the trend earlier: with only $500 million, the maximum gain is 10.62%.
Price already exceeds the news-implied gain: 8.81% versus 5%. The month ends at 5.26%, so a buyer entering at an 8.81% gain loses 3.26%.
One idea to take away
Trends arise from persistent flows and feedback: momentum, positioning, and narratives reinforce one another until fuel runs out.
Record this learning session
Read means only that you confirm reading this chapter. Self-tests are your assessments against reference conclusions. Neither certifies mastery or professional level. Each click retains a timestamped local record.
Answer the core question and record a self-assessment
Why does a price rise often continue for a while?
Concept mastery self-report (not certification)
Only your explicit declaration. Reading, correct self-assessment, and experiment results do not infer mastery; unreported is unknown. Revocation restores unknown and removes selected attachments.
Momentum：Not read
Flow：Not read
Positioning：Not read
Narrative：Not read
Positive feedback：Not read
This chapter's self-report has not been read. No report does not imply mastery.
Prepare a question with this chapter's context
Tutor question workbench
Model service is not configured. Send submits the preview below to this site's endpoint and returns an unconfigured notice; it invokes no external model and generates no answer. Provider, runtime location, and retention remain undecided.
Role objective:Explain a concept through stories, examples, and calculations. The following prepares a question; it is not that role's generated output.
Foundation Chapter 4 · TRD-MKT-004 v1.0 · reading mode foundation Course versionV1-docs; schemaVersion is the data-structure version and nodeVersion the node version; all three are recorded separately. Reading mode is not self-reported Level.
Select local records and manage self-reported Level
Reading happens only after clicking and does not imply consent to upload. Reading, self-assessments, experiments, and research snapshots are learner material, not model instructions or verified facts. Do not include keys, identity details, or real account information.
Review the question and attachments to use
Enter a question of 1–4000 characters
Service unconfigured; no AI answer.
Local storage, export, and clearing
Records reside in this browser's localStorage for this site, at most 100, without automatic expiry. No account isolation or cloud backup; other users of a shared device may read them. Edit in one tab: concurrent writes may overwrite. Export promptly. Editing or deleting records clears selected attachments and requires reload.
Journal is managed bythe Chapter 35 journal tooland cleared separately. Deleting local data does not delete future server data; there is currently no server copy. Storage rejection will not be reported as success. Copy this page's preview.