Our AI Scoring System
How the Scoring Algorithm Works
Believe Capital's proprietary scoring algorithm analyzes new Believe launches across 15 key metrics, generating a quality score between 0-100.
We will use this to detect early projects with potential to invite to AMAs and consider for our own fund. For a small base of holders, we will also provide them with the same technology we'll use to spot potential investment opportunities — early.
The system:
Captures data in real-time as tokens launch
Processes multiple quality indicators simultaneously
Weighs each factor based on historical correlation with quality
Updates scores as new information becomes available
Flags potential concerns or exceptional quality markers
Our algorithm has been trained on thousands of historical Believe launches, with continuous refinement based on market feedback and performance data.
Key Metrics We Analyze
Our scoring system evaluates projects across these primary categories:
Team Assessment
Founder credibility signals
Historical project association
Communication quality
Social presence authenticity
Smart Money Analysis
Early wallet participation patterns
Known quality investor involvement
Capital inflows
Liquidity depth and stability
Engagement Metrics
Authentic vs. artificial engagement
Community growth patterns
Interaction quality
Response dynamics
Market Fit
Narrative alignment
Sector trend correlation
Timing relevance
Differentiation factors
Each category contains multiple sub-metrics that are analyzed and weighted to produce the final quality score.
Detection Methodology
Our detection process follows a systematic approach:
Launch Monitoring
Continuous scanning of new Believe launches
Initial data collection within seconds of launch
Primary Analysis
Evaluation of core metrics
Preliminary quality score assignment
Initial risk assessment
Deep Analysis
Smart money flow evaluation
Team credibility analysis
Engagement pattern analysis
Narrative fit assessment
Quality Score Finalization
Weighted computation of all metrics
Final score assignment (0-100)
Key quality markers identification
Signal Distribution
Format quality detection for Alpha TG
Highlight key quality indicators
Provide contextual information
Example Analysis
Here's an example of our system's analysis process (anonymized):
Project: $MERGE
Initial Data Collection:
Launch timestamp: 2025-05-08 14:32 UTC
Initial liquidity: 82 SOL
Opening market cap: $16,500
Primary Analysis:
Team social accounts: 2+ years old, consistent activity
Launch communication: Clear, well-structured
Initial holder distribution: Healthy, no single dominant wallet
Deep Analysis:
Smart money detection: 3 wallets with strong track records entered within first 60 seconds
Engagement pattern: Organic growth, consistent with quality projects
Narrative alignment: Strong fit with current market interest
No past deploys by creator detected
No previous Twitter name changes
Quality Score: 87/100
Key Quality Markers:
Strong team credibility signals
Healthy smart money participation
Authentic engagement metrics
Solid narrative alignment
This level of detection will be shared in the Alpha TG channels, providing members with an early information advantage.
Last updated