Methodology

Kintechs Reference Portfolio

Portfolio-level metrics are computed on a model allocation that reflects a typical treasury holding of digital assets.

Allocation

50% BTC, 30% ETH, 20% USD stablecoins

Rebalancing

Daily, to constant weights

Prices

Coinbase daily closes, BTC-USD and ETH-USD

Stablecoins

Held at par in the return series; peg risk is measured separately

Value at Risk

VaR estimates the loss the portfolio should not exceed on a given horizon at 99% confidence. Two methods run side by side: the gap between them is information about tail risk and regime change.

Confidence

99%

Historical

1st percentile of daily portfolio returns over the previous 250 trading days, linear interpolation

EWMA

RiskMetrics volatility with λ = 0.94, zero mean, z = 2.326

10-day

1-day VaR scaled by √10

Backtest

A breach is a day whose realised loss exceeds the VaR forecast made with data up to the previous close; compared with the 1% expected rate over 365 days

Expected shortfall and model validation

Expected shortfall

Average loss beyond the 2.5% quantile of daily portfolio returns over 250 days (Basel FRTB standard); 10-day by √10

Backtest

99% historical VaR against realised returns over the last 250 days

Traffic light

Basel zones: green 0–4 breaches, yellow 5–9, red 10 or more

Statistical tests

Kupiec proportion-of-failures and Christoffersen independence tests, p-values reported

Component VaR

Contribution of each asset to portfolio volatility from the EWMA covariance matrix (λ = 0.94)

LVaR

1-day 99% VaR plus the cost of liquidating the BTC and ETH sleeves through the consolidated order book, for portfolios of $10m, $100m and $500m

Stress testing

Historical

The reference portfolio replayed over seven crisis windows from 2020 to 2024 on daily Coinbase closes: total return, peak-to-trough loss, worst day and days to recover the pre-crisis level

Stablecoin shock

In the SVB scenario the stablecoin sleeve follows USDC, which traded at $0.877 on 11 March 2023

Hypothetical

Instantaneous price shocks applied to each sleeve

Reverse

BTC move that produces a given portfolio loss, with ETH moving by its one-year beta to BTC

Volatility and correlation

Realised vol

Sample standard deviation of daily log returns over 30 and 90 days, annualised with √365 (digital assets trade every day)

Correlation

Pearson correlation of daily log returns over 90 days, BTC, ETH and SOL

Performance

Close-to-close price change over 7, 30, 90 and 365 days

Liquidity

A position is worth what it can be sold for. Kintechs prices market sells of $250k to $20m against the live BTC order books of four venues.

Venues

Coinbase (BTC-USD), Kraken (XBT-USD), OKX (BTC-USDT), Binance (BTC-USDT)

Slippage

The sell order consumes bids from the best price down; slippage is the gap between mid and the volume-weighted fill, in basis points

Consolidated

Bids from all venues are expressed relative to each venue's own mid, merged into one book and walked together; this removes the USD/USDT quote difference

Depth

USD value of bids and asks within 1% of mid; imbalance is the bid share of that depth

Spread

Best ask minus best bid, in basis points of mid

Derivatives

Contracts

BTC, ETH and SOL perpetual futures on Kraken Futures

Funding

Hourly relative funding rate averaged per day and annualised (× 24 × 365); charts show a 7-day average

Basis

Mark price against index price, in basis points

Open interest

Open contracts valued at mark price

Stablecoins

Universe

All USD-pegged stablecoins tracked by DefiLlama

Concentration

Herfindahl-Hirschman index of issuer supply shares (0 to 10 000; above 2 500 is highly concentrated)

Chains

Circulating supply aggregated by blockchain

Peg

Kraken order book mid against $1.00, in basis points; DefiLlama price where no Kraken USD market exists

Exchange concentration

Reserves

Assets in wallets disclosed by centralised exchanges, as aggregated by DefiLlama

Share and HHI

Share of the reporting set per exchange; HHI across all reporting exchanges

Change

7-day change in reported reserves, which includes both flows and price moves

Monitoring flags and macro

Stablecoins

Alert: peg deviation above 50 bp or 30-day supply decline above 10%. Watch: above 20 bp or 3%

Exchanges

Alert: reported reserves down more than 10% over 7 days or 5% over 1 day. Watch: 5% or 2.5%

Leverage

Aggregate BTC perpetual open interest on monitored venues as a share of BTC market capitalisation

Macro

Rolling correlation of daily BTC log returns with S&P 500 and broad dollar index log returns and with daily changes in the 10-year Treasury yield (FRED)

On-chain

Hash rate, active addresses and transaction counts from Coin Metrics Community data

Data and schedule

Spot prices

Coinbase Exchange, daily candles

Order books

Coinbase, Kraken, OKX, Binance public market data

Derivatives

Kraken Futures tickers and historical funding

Supply, reserves

DefiLlama stablecoin and CEX transparency data

Schedule

Recalculated daily; the weekly brief is generated from the latest run

Resilience

A source that fails keeps its last good values; see the status page

Stack

Node.js data pipeline, static delivery on Cloudflare