Believe Capital
  • Getting Started
    • Welcome
    • Our mission
  • Basics
    • Accessing our Private Groups
    • The Believe Challenge
    • Believe Capital Fund
    • Our AI Scoring System
    • Private Telegram
    • Founder Discovery
    • Investor Playbooks
    • Founder Playbooks
  • FAQ
  • Terms
Powered by GitBook
On this page
  • How the Scoring Algorithm Works
  • Key Metrics We Analyze
  • Detection Methodology
  • Example Analysis
  1. Basics

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:

  1. Launch Monitoring

    • Continuous scanning of new Believe launches

    • Initial data collection within seconds of launch

  2. Primary Analysis

    • Evaluation of core metrics

    • Preliminary quality score assignment

    • Initial risk assessment

  3. Deep Analysis

    • Smart money flow evaluation

    • Team credibility analysis

    • Engagement pattern analysis

    • Narrative fit assessment

  4. Quality Score Finalization

    • Weighted computation of all metrics

    • Final score assignment (0-100)

    • Key quality markers identification

  5. 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.

PreviousBelieve Capital FundNextPrivate Telegram

Last updated 15 days ago