The Most Reliable Edge in Options¶
There is a structural feature of options markets that has persisted for decades. Implied volatility, the market’s forecast of future movement, almost always overstates what actually happens. The gap between what the market expects and what the market delivers is called the volatility risk premium (VRP).
This is not a theory. It is an empirical fact documented across asset classes, geographies, and time periods. Carr and Wu formalized the framework in 2009, but practitioners had been exploiting the spread between implied and realized volatility long before that.
The VRP exists because options buyers are, on average, paying for insurance. And insurance, by design, costs more than the expected loss. Every premium-selling strategy, from covered calls to iron condors to variance swaps, is ultimately a bet that the VRP will hold.
The Data Has Always Been There¶
The inputs needed to compute the VRP have been publicly available for decades.
Implied volatility is published continuously. The VIX is the most widely quoted volatility measure in the world. You cannot open a trading app without seeing it.
Realized volatility is just math applied to price history. Take a window of returns, compute the standard deviation, annualize it.
The VRP is the difference: implied minus realized. That is it.
And yet, if you open any major brokerage platform today, you will find implied volatility on one screen and historical volatility on another. The actual VRP number, the differential that tells you whether options are cheap or expensive right now, almost nobody computes it for you.
Why Nobody Computes It for You¶
Brokerage platforms show IV and HV side by side. Some offer IV Rank, which tells you where current IV sits relative to its own 52-week range. Useful, but it measures implied volatility against itself, not against realized moves.
EOD analytics platforms compute VRP as a daily metric, comparing yesterday’s closing IV to a 20- or 30-day historical volatility. Academically sound, operationally delayed.
Chart-based indicators attempt VRP classification but inherit the limitations of their platform: delayed feeds, daily close-to-close realized volatility, no awareness of market session transitions.
Institutional terminals give you everything, but you build it yourself. If you have the budget and a quant on staff, you can construct whatever you want. Most traders do not.
The gap is structural. Most analytics platforms are built to scan across thousands of tickers (find the stocks with the highest IV rank, the most unusual flow). VRP monitoring is the opposite: depth on a single instrument, maintaining a live tick buffer, computing rolling variance, and handling every edge case around market transitions. These are fundamentally different architectures, and most platforms are built for the first model.
What Real-Time VRP Requires¶
The implied side is straightforward: read the VIX. The realized side is the hard part.
Traditional realized volatility uses 20 or 30 days of closing prices. It tells you what volatility has been over the past month, not what it is doing right now. Consider a day where the S&P 500 opens flat after a calm week. The 20-day HV reads 12%. But if the market starts moving sharply at 10 AM, intraday RV could spike to 25% within an hour while the daily number barely moves. The regime has changed, but a daily metric will not reflect it until tomorrow.
Real-time VRP requires:
| Component | Purpose |
|---|---|
| Live price feed | Sub-minute observations of the underlying |
| Rolling return buffer | Window of recent log returns for variance calculation |
| Historical baseline | Multi-year distribution for context |
| Regime classifier | Converts raw VRP into an actionable signal |
| Session awareness | Handles market open/close, stale data, weekends |
That last row is the one nobody talks about. Price APIs that return “last known value” outside market hours inject stale observations all weekend. If those are used for variance, you get RV = 0 and a false “elevated VRP” signal at Monday’s open.
What Skavinski Built¶
The Skavinski dashboard includes a VRP widget that runs during regular trading hours:
- 5-second updates. Live SPX and VIX data, recomputed every cycle. Not a 15-minute chart or a daily bar.
- 30-minute rolling realized volatility. Computed from sub-minute price observations, not daily closes. Captures intraday regime shifts as they happen.
- Z-score regime classification. Raw VRP compared against a 735-day historical baseline. Three zones calibrated to the actual distribution, not arbitrary thresholds.
- Session edge handling. Stale data labeled as stale. Rolling buffer flushes on market open so Monday does not inherit Friday.
- Co-located with GEX and expected moves. VRP alongside gamma exposure and expected move ranges in one view.
Why This Matters for Premium Sellers¶
Every credit spread, every iron condor, every short strangle is a bet that implied volatility overstates realized. Knowing the VRP regime in real time changes how you size and manage positions:
- Elevated VRP: Lean into premium collection. The market is paying you more than the expected move warrants.
- Normal VRP: Standard sizing. The edge is present but not exceptional.
- Compressed/inverted VRP: Reduce size or sit out. Premium sellers are being underpaid for the risk.
Most traders approximate this by glancing at the VIX and comparing it to “how the market feels.” The VRP widget replaces intuition with measurement.
The VRP widget is available on the Skavinski Dashboard. It runs during regular trading hours (9:30 AM to 4:00 PM ET) and displays the current regime, z-score, and implied and realized volatility readings.