Worked with · 05 of 12Derivativesderibit.com
Deribit
I collaborated with the Deribit team on research around crypto derivatives market structure, volatility and risk.
Strikes and expiries
A large part of my work was understanding how options markets behave across different strikes and expiries. On Deribit, options can be quoted directly in implied volatility, which makes volatility itself part of the market rather than just an input to a pricing model. I spent time looking at how changes in IV, skew and the term structure affect the price and risk of a position even when the underlying asset has not moved very far.
Fig. 1Volatility across strikes and expiries
Schematic, not data
Direction and volatility
I was particularly interested in the difference between directional exposure and volatility exposure. Two positions can have similar delta at one point in time and behave very differently once gamma, vega, time decay and changes in the volatility surface start to matter. That made it more useful to look at the portfolio as a set of interacting risks rather than a collection of individual contracts.
Fig. 2Same delta, different position
Black-Scholes illustration: 30 days, 60% implied volatility
Margin
Margin was another important part of the work. Deribit’s portfolio margin system stress tests the portfolio against changes in both underlying price and volatility and uses the worst scenario to determine margin requirements. I looked at how hedged portfolios can become more capital efficient, but also at cases where a position that appears well hedged under normal conditions can consume much more margin when volatility rises, correlations change or one side of a hedge becomes less effective.
Fig. 3Margin from the worst scenario
| Volatility change | −12% | −8% | −4% | 0% | +4% | +8% | +12% |
|---|---|---|---|---|---|---|---|
| IV +30 pts | −7.9 | −7.0 | −6.7 | −6.8 | −7.5 | −8.6 | −10.1 |
| IV +15 pts | −5.1 | −3.9 | −3.4 | −3.4 | −4.0 | −5.2 | −6.9 |
| IV 0, unchanged | −2.6 | −1.0 | −0.1 | +0.0 | −0.6 | −1.9 | −3.9 |
| IV −15 pts | −0.3 | +1.8 | +3.1 | +3.4 | +2.7 | +1.2 | −1.1 |
Profit and loss, % of the underlyingWorst cell: -10.1%, the margin
Black-Scholes illustration, not Deribit’s parameters
Perpetuals and futures
I also spent time on perpetuals and futures. Funding creates a continuous transfer between longs and shorts depending on the premium between the perpetual mark price and the index. That makes the basis itself useful information. I looked at how funding, futures curves and options pricing can reinforce or contradict each other, especially around periods of leverage buildup or abrupt changes in positioning.
Fig. 4Three readings of positioning
- FundingThe perpetual’s premium to the index
- Futures curveThe basis by expiry
- OptionsImplied volatility and skew
Execution
Execution quality was another part of the research. A derivatives position can look attractive at mid price or model value while being much harder to execute at size. I looked at order book depth, spread, implied-volatility quotes and the difference between screen liquidity and executable liquidity for larger structures.
For larger option trades, that distinction becomes especially important because institutional execution can move away from the public order book into block trades and RFQs. Deribit allows multi-leg structures to be quoted privately by market makers and then reported back to the exchange. I was interested in how that changes the way liquidity should be measured. Visible order-book depth is only one part of the real market.
Fig. 5Two kinds of liquidity
Screen liquidity
- Order book depth
- Spread
- Implied-volatility quotes
Executable liquidity
- Block trades
- RFQs for multi-leg structures
- Quoted privately, reported to the exchange
Mark price
Mark pricing and liquidation mechanics were also relevant. Deribit does not simply use the last trade for risk calculations. Its mark price uses several market inputs and safeguards designed to reduce the effect of temporary distortions or thin books. In portfolio margin accounts, liquidation can also involve delta hedging through futures or perpetuals rather than simply closing every position independently.
Volatility as a market
I also looked at volatility as its own tradable market. Deribit’s DVOL index and DVOL futures make it possible to separate a view on volatility from a direct directional view on BTC. That creates another way to think about positioning: not just whether BTC moves up or down, but whether the market is underpricing or overpricing the size of the move.
Fig. 6DVOL, a year of daily closes
BTC DVOLMedian 42.8 · range 33.6 to 82.6 · latest 36.5
Monthly averages as a table
| Month | BTC DVOL |
|---|---|
| October 2025 | 44.4 |
| November 2025 | 49.1 |
| December 2025 | 45.6 |
| January 2026 | 41.2 |
| February 2026 | 54.4 |
| March 2026 | 54.3 |
| April 2026 | 44.0 |
| May 2026 | 38.2 |
| June 2026 | 43.5 |
| July 2026 | 37.6 |
| August 2026 | 37.3 |
| September 2026 | 37.2 |
| October 2026 | 35.8 |
Data: Deribit public API, Oct 5, 2025 to Oct 4, 2026
Together
The broader question throughout the collaboration was how much risk a derivatives position actually carries once price, volatility, funding, liquidity and margin are considered together. That was more useful to me than looking at option premium or leverage in isolation.
Fig. 7Considered together
- Price
- Volatility
- Funding
- Liquidity
- Margin
Materials
Public pages about the company and the programs around this work.
Deribit support Portfolio margin Deribit’s own article on portfolio margin, the system that sets margin for a whole portfolio of options and futures.Further reading