🤝Social Trust Scoring

LendFriend uses social collateral—your network on Farcaster—as proof of creditworthiness. When friends contribute to your loan, they vouch for you with both money and reputation.

---

How It Works

1. You request a loan and share it to your network

2. Friends contribute - their backing signals you're trustworthy

3. We measure connection strength using shared mutual friends

4. Lenders see trust scores and decide whether to fund

Close friends with small networks carry more weight than distant followers of influencers. Someone with 20 selective friends signals stronger trust than a stranger with 10,000 connections.

---

Why This Model Works

Research on Prosper.com validates capital-backed social trust:

• Friend bids (capital contributions) reduce defaults by 14% [12]

• Capital-backed endorsements outperform empty endorsements [16]

• Lenders using social + financial data predict defaults 45% better than credit scores alone [12]

The key: friends must risk their own money, not just vouch with words.

This is why LendFriend requires capital contributions. Empty endorsements ("I trust this person") don't predict repayment—only capital-backed vouching works.

---

The Algorithm

We use Adamic-Adar Index to weight connections by how selective they are:

Core insight: A mutual friend with 20 connections is a stronger signal than one with 20,000.

For each mutual connection:
  weight = 1 / log(total_connections)

Trust Score = Σ weights × quality_scores

Research shows Adamic-Adar is 82% more accurate than simple friend counting [7].

The specific thresholds and bonuses will be refined as we collect repayment data. Phase 0 focuses on gathering behavioral data to understand which trust signals best predict repayment.

→ View technical details

---

Risk Tiers

Every lender-borrower pair gets a trust score (0-100) and risk classification:

Risk Level	Trust Score	Meaning
🟢 LOW RISK	≥60	Close social ties, tight-knit community
🟡 MEDIUM RISK	30-59	Some shared connections
🔴 HIGH RISK	<30	Few or no mutual connections

Loans also show Support Strength - what percentage of lenders know the borrower.

---

How Loans Get Funded

Social trust creates natural funding progression:

1. Close friends fund first (small amounts, high trust)

2. Extended network sees validation (joins in)

3. Strangers feel safe (validated by early lenders)

This mirrors traditional lending - ask family first, then friends, then institutions.

---

Anti-Gaming

Creating fake accounts and gaming trust scores is a natural concern. We use algorithmic defenses (quality filtering, Adamic-Adar weighting) combined with economic incentives (lenders risk their own capital) to resist manipulation.

Key defenses:

• Quality filtering removes bots/spam

• Adamic-Adar penalizes large networks

• Real community overlap required

• Lenders vet borrowers (market filtering)

→ Read full security analysis — Covers attack vectors, defenses, limitations, and our iterative approach

---

Why Adamic-Adar?

Research from Prosper.com shows friend capital contributions reduce defaults by 14% [12].

We chose Adamic-Adar because it mathematically captures what matters:

• Selective connections count more than large follower lists

• Mutual close friends signal stronger accountability

• Fast to compute from partial social graph data

The algorithm was designed for link prediction, not lending - we're testing whether it translates to repayment prediction in Phase 0.

---

Next: The Algorithm · Risk Scoring · Research