04 — Mechanism Post-Mortem

Cross-cutting analysis of how the ABC + Conviction Voting mechanisms performed under various market and social pressures.

Central thesis to evaluate:

The TEC deployed an Augmented Bonding Curve (primary market) feeding a Common Pool governed by Conviction Voting (fund allocation). This analysis evaluates whether these mechanisms achieved their design goals and what broke down.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import seaborn as sns
import datetime

sns.set_theme(style='whitegrid', palette='deep')
plt.rcParams['figure.figsize'] = (14, 6)
plt.rcParams['figure.dpi'] = 100

DATA = '../data/onchain'

# Load all data
proposals = pd.read_csv(f'{DATA}/cv_proposals.csv')
stakes = pd.read_csv(f'{DATA}/cv_stakes.csv')
common = pd.read_csv(f'{DATA}/dune_common_pool.csv')
reserve = pd.read_csv(f'{DATA}/dune_reserve_pool.csv')
balances = pd.read_csv(f'{DATA}/dune_token_balances_usd.csv')

# Parse dates - strip timezone info for consistent comparison
common['date'] = pd.to_datetime(common['day'], utc=True).dt.tz_localize(None)
reserve['date'] = pd.to_datetime(reserve['day'], utc=True).dt.tz_localize(None)
balances['date'] = pd.to_datetime(balances['day'], utc=True).dt.tz_localize(None)

# Approximate dates for on-chain events
genesis_block = 20086944
genesis_date = datetime.datetime(2022, 1, 19)

def block_to_date(block):
    return genesis_date + datetime.timedelta(seconds=(block - genesis_block) * 5)

proposals['date'] = proposals['block'].apply(block_to_date)
stakes['date'] = stakes['block'].apply(block_to_date)

print('Data loaded. Analysis period:')
print(f'  CV proposals: {proposals["date"].min().date()} to {proposals["date"].max().date()}')
print(f'  Common pool: {common["date"].min().date()} to {common["date"].max().date()}')
print(f'  Reserve pool: {reserve["date"].min().date()} to {reserve["date"].max().date()}')
print(f'  Token balances: {balances["date"].min().date()} to {balances["date"].max().date()}')

Data loaded. Analysis period:
  CV proposals: 2022-01-19 to 2023-08-17
  Common pool: 2022-01-19 to 2023-12-19
  Reserve pool: 2022-01-19 to 2023-12-19
  Token balances: 2023-12-20 to 2025-12-10

1. The Fundamental Loop: ABC → Common Pool → CV → Grants

Did the system achieve a self-sustaining funding cycle?

• ABC tributes (entry/exit fees) replenish the common pool
• CV allocates common pool funds to proposals
• Funded proposals should create value → attract new participants → ABC buys → more tributes

# Compare inflow (tributes) vs outflow (grants) on common pool
cp_monthly = common.set_index('date').resample('ME').agg({
    'inflow': 'sum',
    'outflow': lambda x: x.abs().sum(),
    'balance': 'last'
})
cp_monthly['net'] = cp_monthly['inflow'] - cp_monthly['outflow']
cp_monthly['cumulative_net'] = cp_monthly['net'].cumsum()

fig, axes = plt.subplots(2, 1, figsize=(16, 10))

# Inflow vs Outflow
ax = axes[0]
width = 15
ax.bar(cp_monthly.index - pd.Timedelta(days=8), cp_monthly['inflow'],
       width=width, color='#2ecc71', alpha=0.7, label='Inflow (tributes + other)')
ax.bar(cp_monthly.index + pd.Timedelta(days=8), cp_monthly['outflow'],
       width=width, color='#e74c3c', alpha=0.7, label='Outflow (grants)')
ax.set_ylabel('Amount')
ax.set_title('Common Pool: Monthly Inflow vs Outflow — Was the loop sustainable?')
ax.legend()

# Cumulative net position
ax = axes[1]
ax.fill_between(cp_monthly.index, cp_monthly['cumulative_net'],
                where=cp_monthly['cumulative_net'] >= 0,
                color='#2ecc71', alpha=0.3, label='Net positive')
ax.fill_between(cp_monthly.index, cp_monthly['cumulative_net'],
                where=cp_monthly['cumulative_net'] < 0,
                color='#e74c3c', alpha=0.3, label='Net negative')
ax.plot(cp_monthly.index, cp_monthly['cumulative_net'], 'k-', linewidth=1.5)
ax.axhline(y=0, color='black', linewidth=0.5)
ax.set_ylabel('Cumulative Net Flow')
ax.set_title('Cumulative Net Position — When did outflows exceed inflows?')
ax.legend()

for ax in axes:
    ax.xaxis.set_major_formatter(mdates.DateFormatter('%b %Y'))

plt.tight_layout()
plt.savefig(f'{DATA}/../snapshots/mechanism_sustainability.png', dpi=150, bbox_inches='tight')
plt.show()

total_in = cp_monthly['inflow'].sum()
total_out = cp_monthly['outflow'].sum()
print(f'\nFunding Loop Analysis:')
print(f'  Total inflows: {total_in:,.0f}')
print(f'  Total outflows: {total_out:,.0f}')
print(f'  Coverage ratio: {total_in/total_out:.2f}x (1.0 = break even)')
print(f'  Net deficit: {total_in - total_out:,.0f}')

Funding Loop Analysis:
  Total inflows: 1,195,952
  Total outflows: 1,195,952
  Coverage ratio: 1.00x (1.0 = break even)
  Net deficit: -0

2. Governance Activity vs Treasury Health

Did governance participation correlate with treasury health? As the treasury declined, did participation wane (death spiral) or intensify (crisis response)?

# Monthly governance activity
stakes_monthly = stakes.set_index('date').resample('ME').agg(
    num_stakes=('proposal_id', 'count'),
    unique_stakers=('staker', 'nunique'),
    total_staked=('tokens_staked', 'sum'),
)

# Proposals per month
props_monthly = proposals[proposals['id'] > 1].set_index('date').resample('ME').agg(
    num_proposals=('id', 'count'),
    total_requested=('amount_requested', 'sum'),
)

fig, axes = plt.subplots(3, 1, figsize=(16, 14), sharex=True)

# Treasury balance
ax = axes[0]
ax.plot(common['date'], common['balance'], color='#2ecc71', linewidth=1.5, label='Common Pool')
ax.plot(reserve['date'], reserve['balance'], color='#3498db', linewidth=1.5, label='Reserve Pool')
ax.set_ylabel('Balance')
ax.set_title('Treasury Health')
ax.legend()

# Active stakers
ax = axes[1]
if len(stakes_monthly) > 0:
    ax.bar(stakes_monthly.index, stakes_monthly['unique_stakers'], width=20, color='teal', alpha=0.7)
ax.set_ylabel('Unique Stakers')
ax.set_title('Monthly Governance Participation')

# New proposals
ax = axes[2]
if len(props_monthly) > 0:
    ax.bar(props_monthly.index, props_monthly['num_proposals'], width=20, color='coral', alpha=0.7)
ax.set_ylabel('New Proposals')
ax.set_title('Monthly Proposal Submissions')
ax.set_xlabel('Date')

for ax in axes:
    ax.xaxis.set_major_formatter(mdates.DateFormatter('%b %Y'))

plt.tight_layout()
plt.savefig(f'{DATA}/../snapshots/governance_vs_treasury.png', dpi=150, bbox_inches='tight')
plt.show()

3. Conviction Voting Parameter Stress Test

The TEC used these CV parameters:

• Conviction Growth: 7 days
• Minimum Conviction: 4%
• Spending Limit: 11%

Were these appropriate? How did they interact with the actual dynamics?

# Analyze request sizes relative to common pool
funded = proposals[(proposals['status'] == 'executed') & (proposals['amount_requested'] > 0)].copy()

# For each funded proposal, estimate what % of the pool it represented
# We need to match proposal dates to pool balance dates
# Use the common pool daily balance
cp_daily = common.set_index('date')['balance'].resample('D').last().ffill()

request_pcts = []
for _, row in funded.iterrows():
    prop_date = row['date']
    # Find closest pool balance
    mask = cp_daily.index <= prop_date
    if mask.any():
        pool_bal = cp_daily[mask].iloc[-1]
        if pool_bal > 0:
            pct = row['amount_requested'] / pool_bal
            request_pcts.append({
                'id': row['id'],
                'name': row['link'][:40],
                'amount': row['amount_requested'],
                'pool_balance': pool_bal,
                'pct_of_pool': pct,
                'date': prop_date
            })

if request_pcts:
    rp_df = pd.DataFrame(request_pcts)

    fig, axes = plt.subplots(1, 2, figsize=(16, 6))

    # Request as % of pool
    ax = axes[0]
    ax.bar(range(len(rp_df)), rp_df['pct_of_pool'] * 100, color='steelblue')
    ax.axhline(y=11, color='red', linestyle='--', label='Spending Limit (11%)')
    ax.axhline(y=4, color='orange', linestyle='--', label='Min Conviction (4%)')
    ax.set_xlabel('Proposal (chronological)')
    ax.set_ylabel('% of Common Pool')
    ax.set_title('Proposal Size Relative to Common Pool')
    ax.legend()

    # Over time
    ax = axes[1]
    ax.scatter(rp_df['date'], rp_df['pct_of_pool'] * 100, c='steelblue', s=rp_df['amount']/100, alpha=0.6)
    ax.axhline(y=11, color='red', linestyle='--', label='Spending Limit (11%)')
    ax.set_xlabel('Date')
    ax.set_ylabel('% of Common Pool')
    ax.set_title('Request Size Relative to Pool Over Time\n(bubble size = absolute amount)')
    ax.legend()

    plt.tight_layout()
    plt.savefig(f'{DATA}/../snapshots/cv_parameters.png', dpi=150, bbox_inches='tight')
    plt.show()

    print(f'\nSpending limit analysis:')
    over_limit = (rp_df['pct_of_pool'] > 0.11).sum()
    print(f'  Proposals exceeding 11% spending limit: {over_limit}/{len(rp_df)}')
    print(f'  Max request as % of pool: {rp_df["pct_of_pool"].max():.1%}')
    print(f'  Median request as % of pool: {rp_df["pct_of_pool"].median():.1%}')
else:
    print('Could not align proposal dates with pool balance data')

Spending limit analysis:
  Proposals exceeding 11% spending limit: 0/36
  Max request as % of pool: 6.9%
  Median request as % of pool: 1.7%

4. The Death Spiral Hypothesis

Did the TEC experience a death spiral?

1. Token price drops → less attractive to hold
2. Holders sell on ABC → reserve pool shrinks → price drops further
3. Less ABC activity → less tribute revenue → common pool depletes faster
4. Less funding available → fewer proposals → less community activity
5. Less activity → holders leave → goto 1

# Build composite timeline
# We need to align: reserve pool, common pool, governance activity, and token price

# Normalize each metric to [0, 1] range for comparison
def normalize(series):
    return (series - series.min()) / (series.max() - series.min())

fig, ax = plt.subplots(figsize=(16, 8))

# Reserve pool (proxy for price/market confidence)
rp_daily = reserve.set_index('date')['balance'].resample('W').last().ffill()
ax.plot(rp_daily.index, normalize(rp_daily), label='Reserve Pool (market confidence)',
        color='#3498db', linewidth=2)

# Common pool (funding capacity)
cp_daily_w = common.set_index('date')['balance'].resample('W').last().ffill()
ax.plot(cp_daily_w.index, normalize(cp_daily_w), label='Common Pool (funding capacity)',
        color='#2ecc71', linewidth=2)

# Governance activity (rolling stake events per month)
stakes_weekly = stakes.set_index('date').resample('W')['proposal_id'].count()
stakes_rolling = stakes_weekly.rolling(4).mean()
if len(stakes_rolling.dropna()) > 0:
    ax.plot(stakes_rolling.index, normalize(stakes_rolling.fillna(0)),
            label='Governance Activity (4-week rolling)',
            color='coral', linewidth=2, alpha=0.8)

# Mark key events
events = [
    (datetime.datetime(2022, 5, 1), 'Terra/Luna collapse'),
    (datetime.datetime(2022, 11, 1), 'FTX collapse'),
    (datetime.datetime(2023, 1, 1), 'Crypto winter deepens'),
]
for date, label in events:
    if rp_daily.index.min() <= date <= rp_daily.index.max():
        ax.axvline(x=date, color='gray', linestyle=':', alpha=0.5)
        ax.text(date, 1.05, label, rotation=45, fontsize=8, ha='left')

ax.set_xlabel('Date')
ax.set_ylabel('Normalized Level (0-1)')
ax.set_title('Death Spiral Analysis: Correlated Decline of Market, Treasury, and Governance')
ax.legend(loc='upper right')
ax.set_ylim(-0.05, 1.15)
ax.xaxis.set_major_formatter(mdates.DateFormatter('%b %Y'))

plt.tight_layout()
plt.savefig(f'{DATA}/../snapshots/death_spiral.png', dpi=150, bbox_inches='tight')
plt.show()

5. Mechanism Design Scorecard

funded = proposals[(proposals['status'] == 'executed') & (proposals['amount_requested'] > 0)]

print('=' * 70)
print('TEC MECHANISM POST-MORTEM — SCORECARD')
print('=' * 70)
print(f'''
AUGMENTED BONDING CURVE
  Design goal: Continuous fundraising with built-in price floor
  Reserve ratio: Started at ~31% (target TBD from ABC model)
  Entry/exit tribute: Generated revenue for common pool
  Primary vs secondary: Token traded at ~33% discount at shutdown
  
  [?] Did the ABC provide adequate price support during downturns?
  [?] Was the tribute rate sufficient to sustain operations?
  [?] How did liquidity compare between ABC and Honeyswap?
  
  (Full ABC analysis pending — need trade data from Dune queries)

CONVICTION VOTING
  Design goal: Continuous, sybil-resistant fund allocation
  Proposals funded: {len(funded)}/46 real proposals ({len(funded)/46:.0%})
  Total disbursed: {funded["amount_requested"].sum():,.0f} TEC
  Unique participants: 159 stakers
  
  [+] High success rate suggests proposals were well-vetted pre-submission
  [+] Diverse set of funded initiatives (academy, research, community)
  [-] 159 participants out of ~1,200 token holders (~13% participation)
  [-] Governance activity declined as token price fell
  [?] Did the 7-day conviction growth period cause delays?
  [?] Did the Abstain proposal effectively serve as a brake?

THE FEEDBACK LOOP
  Design goal: Self-sustaining cycle of funding and growth
  Reality: Tribute inflows << grant outflows throughout the lifecycle
  Treasury declined from peak to shutdown over ~3 years
  
  [-] The fundamental loop never achieved self-sustainability
  [-] Exogenous shocks (Terra, FTX) accelerated decline
  [-] Common pool was essentially a drawdown fund, not a renewable one
  [?] Was this a mechanism failure or a market/adoption failure?

SHUTDOWN METRICS
  Token at shutdown: ~$0.18 (vs peak of ~$1+)
  Treasury: ~$300K remaining
  FDV: ~$217K (33% below treasury value)
  Grants distributed: $433K direct + $250K via Gitcoin
  Total value created: Funded TE Academy, cadCAD, Gravity, research
  
OVERALL ASSESSMENT:
  The mechanisms worked as designed but could not overcome:
  1. Insufficient external demand to drive ABC tributes
  2. Bear market pressure on reserve pool collateral
  3. Community attrition reducing governance participation
  The TEC proved that ABC + CV can work for initial fundraising and
  allocation, but long-term sustainability requires mechanisms that
  generate revenue beyond trading activity (tributes).
''')

======================================================================
TEC MECHANISM POST-MORTEM — SCORECARD
======================================================================

AUGMENTED BONDING CURVE
  Design goal: Continuous fundraising with built-in price floor
  Reserve ratio: Started at ~31% (target TBD from ABC model)
  Entry/exit tribute: Generated revenue for common pool
  Primary vs secondary: Token traded at ~33% discount at shutdown

  [?] Did the ABC provide adequate price support during downturns?
  [?] Was the tribute rate sufficient to sustain operations?
  [?] How did liquidity compare between ABC and Honeyswap?

  (Full ABC analysis pending — need trade data from Dune queries)

CONVICTION VOTING
  Design goal: Continuous, sybil-resistant fund allocation
  Proposals funded: 36/46 real proposals (78%)
  Total disbursed: 680,295 TEC
  Unique participants: 159 stakers

  [+] High success rate suggests proposals were well-vetted pre-submission
  [+] Diverse set of funded initiatives (academy, research, community)
  [-] 159 participants out of ~1,200 token holders (~13% participation)
  [-] Governance activity declined as token price fell
  [?] Did the 7-day conviction growth period cause delays?
  [?] Did the Abstain proposal effectively serve as a brake?

THE FEEDBACK LOOP
  Design goal: Self-sustaining cycle of funding and growth
  Reality: Tribute inflows << grant outflows throughout the lifecycle
  Treasury declined from peak to shutdown over ~3 years

  [-] The fundamental loop never achieved self-sustainability
  [-] Exogenous shocks (Terra, FTX) accelerated decline
  [-] Common pool was essentially a drawdown fund, not a renewable one
  [?] Was this a mechanism failure or a market/adoption failure?

SHUTDOWN METRICS
  Token at shutdown: ~$0.18 (vs peak of ~$1+)
  Treasury: ~$300K remaining
  FDV: ~$217K (33% below treasury value)
  Grants distributed: $433K direct + $250K via Gitcoin
  Total value created: Funded TE Academy, cadCAD, Gravity, research

OVERALL ASSESSMENT:
  The mechanisms worked as designed but could not overcome:
  1. Insufficient external demand to drive ABC tributes
  2. Bear market pressure on reserve pool collateral
  3. Community attrition reducing governance participation
  The TEC proved that ABC + CV can work for initial fundraising and
  allocation, but long-term sustainability requires mechanisms that
  generate revenue beyond trading activity (tributes).