Hi everyone,
This is a short research note I’m sharing for discussion and feedback.
It is not a proposal, not a grant request, and not a recommendation, just an exploratory analysis that may be useful for thinking about risk dynamics around Lido-related activity.
My background is in statistics and time series analysis, and over the last months I’ve been looking at volatility regimes and large-flow behavior across DeFi protocols.
What I looked at
At a high level, the analysis focuses on:
-
Volatility clustering using rolling-window estimates (conditional volatility)
-
Simple regime classification (low vs. high volatility states)
-
Periods of large ETH movements interacting with Lido and downstream protocols (e.g. lending venues)
The goal was not prediction, but to explore whether certain patterns tend to repeat during stress or reallocation phases.
Main observations (very preliminary)
Some early patterns that stood out:
-
Volatility tends to cluster around large redeployment events rather than isolated price moves.
-
High-volatility regimes appear to coincide with periods of reduced retention and higher cross-protocol activity.
-
From a risk perspective, these regimes seem to matter more for leverage and secondary usage than for staking itself.
Again, these are observations, not conclusions.
Why this might matter
A regime-based view of risk could be useful for:
-
Contextualizing large withdrawals or redeployments
-
Stress-testing assumptions around downstream usage of stETH
-
Informing future monitoring or research dashboards
I’m mainly interested in whether this framing resonates with how others think about risk at the DAO or protocol level.
Open questions
-
Does this type of regime-based view align with how risk is discussed internally?
-
Are there existing analyses or dashboards that already approach this problem in a similar way?
-
Would it make sense to extend this toward more formal stress indicators?
Happy to clarify methodology or share more details if useful.
Thanks for reading.
