Do you remember blindly believing in the 'big shots' signals when you first entered the circle? I followed signals in a paid community for three months, losing 10 ETH, and ultimately became disheartened due to the inconsistent quality of information sources. It wasn't until I started to deeply use Rumour.app, quantifying and tracking the relationship between its rating system and market prices, that I discovered—market sentiment and the credibility of information can be captured through data models.
Technical architecture: Deconstructing rumors like analyzing financial reports
As a data analyst, what attracts me most about Rumour.app is its rating system. It transforms vague 'rumors' into quantifiable data metrics, primarily consisting of three core modules:
Initial credibility:
Automatically generated based on publisher's historical accuracy rate, account weight, etc.
A core contributor with a historical accuracy rate of 80% can have an initial score of up to 6.5.
Verification weight:
Integrates various verification methods such as on-chain data, code submissions, media links, etc.
The weight of an on-chain large transfer evidence is far higher than a Twitter screenshot.
User verification providing counter-evidence will negatively impact the score.
Time decay:
Rumor scores will naturally decay over time, unless new, strong evidence emerges.
This effectively avoids the continuous misguidance of outdated information to users.
The decay coefficient is determined by the category of the rumor (e.g., short-term price, long-term collaboration).
Real data: why high-scoring rumors are worth paying attention to
According to Dune Analytics dashboard #9133, the backtest of data from the past six months shows:
Total number of verified rumors: 4,821
Number of rumors with scores above 8.0: 315
Average positive fulfillment rate of rumors above 8.0 within 72 hours: 78.6%
Average community follow-on return rate (simulated): 23.5%
These data are public, proving the effectiveness of the scoring system in screening high-value information.
Cost comparison: traditional information analysis vs Rumour.app
I compared the costs and efficiencies of two information acquisition methods:
Traditional method (manual screening):
Tool subscription fee: average $350 per month (Nansen, Arkham, etc.)
Time cost: average 3-4 hours per day
Information noise ratio: about 80%
Rumour.app method:
Time cost: average 0.5-1 hour per day
Core information screening efficiency improvement: about 85%
Opportunity cost savings: immeasurable, but successfully avoided multiple false information traps
Ecosystem case: $ALT's accurate prediction
Well-known trader @Crypto_Sage shared his experience using Rumour.app:
Before $ALT went live, he noticed a rumor about 'Binance Labs additional investment', with an initial score of only 5.8.
Later, several high-weight users submitted on-chain address association evidence, and the score rose to 8.2 within 6 hours.
Based on this, he positioned himself in advance and ultimately gained over 200% profit.
"The scoring curve of Rumour.app feels like a leading 'sentiment indicator' for me, ahead of the candlestick chart." @Crypto_Sage summarized this in the community.
Token economy: value capture of $RUM
$RUM (hypothetical token) plays a core role in the scoring system:
Information staking: Publishing high-value rumors requires staking $RUM, and false information will be confiscated.
Verification incentives: Users who successfully verify or falsify rumors will receive $RUM rewards.
Data access: Viewing in-depth scoring data analysis (such as verifier profiles, fund flows) requires consuming $RUM.
Risks and challenges
Of course, the scoring system is not perfect:
There is a risk of 'score manipulation' and 'group verification'.
For macro or long-term rumors, the difficulty of quantitative scoring is greater.
The community needs to be vigilant against 'scoreism' while ignoring independent judgment.
Interactive discussion: your information decision-making basis
"What type of information do you value most when making trading decisions? Is it on-chain data, community sentiment, or technical analysis? Share your methods, and we will select three of the most profound respondents to give one month of premium data access to Rumour.app."
