We've now built the conceptual framework — the contract, pricing, the Greeks, volatility, expected value, and edge. None of that matters, however, if you cannot translate it into actual execution. This section moves from theory into the real world: real option chains, real brokerage workflow, order placement, liquidity, margin, position sizing, and how to actually evaluate and place trades in practice. We'll work through real examples and show how to analyze trades structured around the variance risk premium. This is where the framework stops being theory and becomes a trading process.
For this purpose, I set up an Interactive Brokers account with approximately $110,000 — roughly $100,000 USD and $14,000 CAD. The plan is to use it for the trades shown in this section and for future real-trading content. If there is anything specific you'd like to see covered in those videos, let me know in the comments.
Because I am located in Canada, all overall account figures will be shown in Canadian dollars. Any margin numbers displayed at the top of the platform — and any margin quotes for individual trades — will be in CAD, which at the current exchange rate is worth approximately 0.72 USD.
Net liquidity, also called net liquidation value, is the mark-to-market value of the account if all positions were liquidated at current prices. With no positions open, NLV is simply the Canadian-dollar equivalent of cash held in the account.
Maintenance margin is the minimum equity required to continue holding the current positions. As positions are opened, this figure changes to reflect the risk of those positions.
Excess liquidity is the margin buffer. At Interactive Brokers, it is best thought of as equity with loan value minus maintenance margin. In a simple account with no borrowing, it will often feel similar to the difference between NLV and maintenance margin, but the exact broker definition matters. When excess liquidity falls too low or turns negative, positions may be liquidated.
These numbers move continuously. They change as prices move, as volatility changes, and as the broker adjusts its internal risk requirements. For Canadian residents at Interactive Brokers, the account operates under a rule-based margin framework — portfolio margin is not currently available. That generally makes the account less forgiving than a true risk-based portfolio margin regime.
On top of that, losses on positions reduce NLV, which also reduces excess liquidity and shrinks the margin buffer. Brokers can also raise house margin requirements at their discretion, often without notice, which is why margin must be monitored continuously.
These guidelines are not broker rules, but practical risk-management heuristics. The appropriate level depends heavily on how hedged the overall portfolio is.
Margin on positions matters significantly for capital efficiency and overall returns. Consider two trades with the same theoretical edge: one requires $10,000 in maintenance margin, the other only $2,000. All else equal, the $2,000 margin position would show a 5× higher return on margin and is therefore more capital efficient.
That said, return on margin should not be evaluated in isolation. We still need to consider the absolute risk of the trade, the tail risk, the liquidity, and how the position fits into the overall portfolio. A lower-margin trade is not automatically better — but when two trades offer similar edge and similar risk, the one requiring less margin is generally the more capital-efficient choice. We'll see this play out shortly using real positions.
For these features, the universe is filtered to all ETFs and stocks with:
Three features stand out:
For more on the research behind these features. It's important to remember these are generalities — each potential trade and its underlying situation must still be assessed before taking it.
Using these features and a model, we've curated a screener on the OQuants homepage that surfaces tickers meeting the criteria. We'll start by looking at IBIT.
IBIT tracks Bitcoin's spot price. Recently, Bitcoin has fallen meaningfully, but over the past month or two it has also chopped sideways.
Looking at the term structure, RV — recent realized volatility — is falling sharply, while IV remains relatively high and sticky. Stripping out RV, the term structure is clearly upward sloping and in contango, generally a good predictor of future short-volatility returns.
The IV percentile vs. VRP regression shows that the higher the IV percentile, the worse the VRP. Since we're relatively high in percentile here, that's a mildly contradictory signal for short-vol returns. However, the most recent 10 days have clustered in a positive VRP range.
Examining the realized VRP chart — IV at the time vs. subsequent realized volatility — the current ratio is around 20%, with a mean around 15%. So blindly selling volatility on IBIT would have been profitable historically.
The non-lagged ratio, current IV over current-period RV, is around 11%, with a mean near 15%.
Both readings suggest a positive VRP on this trade in general. Combined with the relatively high, sticky IV and the falling RV in the term structure, there appears to be a strong opportunity to sell volatility on IBIT.
Taking that as the baseline edge, we can price the position. Starting with the at-the-money straddle, we find the expiration nearest 30 days and sell ATM contracts — those closest to 50 delta, or the strike nearest the underlying price of approximately 39.
The resulting short straddle implies a current IV around 52%. Using the past 30-day RV of approximately 45% as our forecast, the scenario analysis returns:
In practice, these Kelly fractions are far larger than what we'd actually want to trade. We'll address sizing shortly.
The platform also provides an effective fill IV calculator, which takes a spread price and returns the IV you actually sold. This is critical when placing orders — it prevents you from chasing fills, as we'll see.
To build a covered position, we can construct an iron condor:
The reason to buy cheap wings is to avoid eating into our edge too much. Using the same baseline RV forecast of 45% and running the scenario analysis, the estimated edge is approximately 11.6% — slightly less than the straddle. The reduction comes from financing the protection by purchasing the OTM wings.
When we're taking short-volatility trades, our view is that the implied volatility of the market is overpriced. So when we purchase these wing contracts, we are implicitly overpaying — because our thesis is that volatility is too high, and yet we are still buying volatility in those wings.
A second factor is volatility skew. Normalized by call delta, OTM put IV is considerably higher than OTM call IV. By buying the wings — particularly the put wing — we are paying a substantial premium in IV for protection, while selling near the 50 delta where IV is significantly lower. That asymmetry also eats into edge.
A few situations would trigger an early exit:
GME also showed up on the stock VRP screener.
The IV percentile vs. VRP regression shows GME has a fairly consistent VRP regardless of IV percentile — positive at both low and high readings, and slightly larger at higher IV percentiles. Recent points cluster above the regression line, in the positive VRP range.
Examining the IV vs. lagged RV — the actual historical spread — the mean is around 18% and the last value is around 46%. There is a consistently high VRP on GME.
The non-lagged version, IV over last-period RV, has a mean around 14% and a current value of around 16%.
The term structure is broadly contango and upward sloping, especially from the 30-day baseline we'll look to sell. RV is falling, with a considerable spread between last-period RV and current IV. All signals point to a good short-vol opportunity.
Skew is present in both directions. On the call side in particular, GME is a heavily retail-interested name — most viewers will remember the run it made a few years ago — and retail traders consistently bid up and buy out-of-the-money call options, pushing call skew upward.
As a general rule: in any heavily retail-interested name with price-insensitive option buyers, there is usually a good short-vol opportunity.
The downside of that skew is that, when building a static structure, we pay considerably more for wing protection than we receive selling near the center.
The 30-day expiration with underlying around 22.50 gives us a 50-delta strike near 23. Using the 30-day most recent realized volatility as our forecast — 32.5% — the scenario analysis returns:
Again, far higher than the position size we'd actually take.
A static structure on GME:
Comparing this to the IBIT iron condor, the risk-reward looks much less attractive — and the reason is the amount paid for wings due to skew. With our baseline RV assumption of 32.5%, the scenario analysis returns a mean return around 10%, lower than the straddle because of the wings we're financing.
As with IBIT, we exit early if:
Drift is again the enemy. Even if the realized volatility on the path is less than what we sold, sustained directional drift can still produce a loss.
The Kelly fractions returned by the scenario analyses are far too large to trade at face value. There are two reasons.
First, full Kelly is intrinsically uncomfortable. Kelly's objective is to maximize terminal wealth. It is unconcerned with the path your P&L takes — only the final value. As a result, betting full Kelly produces large portfolio variance, with 50%+ drawdowns being acceptable so long as the terminal value is high. For most people this is an unsustainable equity curve. I have other videos on the channel exploring P&L paths and drawdown statistics under full Kelly.
Second, the underlying simulation assumes normality. The scenario analysis generates price paths via geometric Brownian motion, which embeds a normality assumption. As covered in earlier sections, real return distributions have fat tails and negative skew — they can move much further than implied by a normal distribution.
A reasonable rule of thumb is to use 5% to 10% of the suggested Kelly fraction for VRP or short volatility trades. Taking the conservative 5% on the IBIT straddle:
5% × 30% = 1.5%
We want to sell roughly 1.5% of account value in premium for this straddle. With an account value of $110,000, that comes to approximately $1,700 in premium, which at the current straddle fill price works out to about four lots.
For consistency and comparison purposes — and to illustrate the structural differences — we'll trade four lots of each of the four positions (the IBIT straddle, the IBIT iron condor, the GME straddle, and the GME iron condor). In reality, trading four lots of the straddle plus four lots of the condor on the same underlying is effectively doubling up the same short-volatility view; this is shown here only to highlight the differences between covered (condor) and uncovered (straddle) structures, especially as delta hedging comes into play.
We're on the Interactive Brokers platform, working with the same account shown previously, with the GME options chain open at the April 24th, 29 DTE expiration.
GME SpreadsBuilding the straddle priced out earlier on OQuants, the first thing to notice is a bid-ask spread roughly 40 cents wide. For options, a 40-cent spread is not unusual. It does not mean we'll have to sell at the bid for the smallest credit (e.g., –$1.79 per lot) — we can work the order book toward a better fill.
The margin impact for four lots of the straddle is around $25,000 CAD. Opening the margin window shows both:
All else equal, it is common for initial margin to be roughly 2× maintenance margin. This is why the 50%-maintenance-margin guideline matters: maintenance margin can easily jump toward initial margin during a volatility shock or after a rule-based broker margin hike.
Next, we build the four-lot iron condor. The margin impact is dramatically smaller — approximately $3,000 CAD versus the $25,000 for the straddle. This is important: even though buying wings gives up some volatility edge, the reduction in margin impact (and corresponding increase in return on margin) can more than compensate, depending on your margining regime.
Both GME spreads are placed near the bid — unlikely to fill immediately. We'll walk them down shortly.
IBIT SpreadsBuilding the IBIT short straddle, two things stand out:
IBIT has substantially more option volume and open interest, and is quoted more tightly than GME. IBIT is the more liquid name.GME straddle, despite selling significantly more premium on a higher-priced underlying. Margin efficiency — and potential edge per dollar of margin — is therefore higher on IBIT.The four-lot IBIT iron condor again shows a much tighter spread than the GME condor. Its margin is slightly higher than the GME condor, because covered structures are margined differently than uncovered ones.
Both IBIT orders are also submitted at the bid and need to be worked.
We are far more likely to fill near the mid than at the bid, so we walk these orders down — decreasing the credit received by a couple of cents at a time — until they fill. Final fills:
| Position | Fill (Credit per spread) |
|---|---|
GME Iron Condor | $0.99 |
IBIT Iron Condor | $2.34 |
IBIT Short Straddle | $4.50 |
GME Short Straddle | $1.99 |
One typical observation when entering option spread trades: upon fill, the instantaneous unrealized P&L is almost always negative. This is due to wide option spreads and the mark being placed near the mid — so by the time you fill, your unrealized P&L at the moment of entry is generally slightly negative.
Looking at the account after execution: maintenance margin is approximately $30,000 CAD, excess liquidity is approximately $128,700 CAD, and maintenance as a fraction of NLV is around 20% — well within the recommended range.
Reminder: in reality we would never simultaneously open an iron condor and a short straddle on the same name. This is for illustration only — most of the maintenance margin here is being consumed by the uncovered short straddles. Under rule-based margining (and not portfolio margin), giving up some edge to use a covered structure like an iron condor or iron butterfly can be significantly advantageous from a margin perspective.
How did we know what price to walk the order to — and when to stop? We didn't. That's what the effective fill IV calculator is for, and it's something you should always check before walking orders.
GME Iron CondorFilled at $0.99 credit per spread, the effective fill IV was 35.23%. The 30-day at-the-money market IV was around 38.5%. By walking the order down and buying the wings, we effectively sold an IV of 35.23%. With our forecast RV at 32.5%, this leaves only about 3% of IV edge in the trade.
GME Short StraddleFilled at $1.99 credit per spread, the effective fill IV is approximately 38.5% — substantially higher than the condor. This is the tradeoff between an uncovered structure (straddle/strangle, which typically gives up less edge) and a covered one (condor, which uses far less margin).
If wide spreads tempt you into walking the GME condor down further — say, to $0.85 of credit — the effective fill IV becomes 32.2%, which is less than the RV we used to benchmark the trade. At that fill, you've implicitly entered a trade with a negative IV-vs-RV-forecast spread simply by chasing price. Always check effective fill IV before continuing to walk an order.
IBIT Iron CondorFilled at $2.34 credit per spread, the effective fill IV is approximately 49%. Against the benchmark RV forecast of 44.5%, we're still well within the range to capture meaningful edge.
IBIT Short StraddleFilled at $4.50 credit per spread, the effective fill IV is significantly higher. Same tradeoff: with an uncovered structure we sell a higher effective IV and theoretically capture more edge, at the cost of substantially more margin.
The takeaway is that you must always evaluate both margin and edge for each fill, then judge which structure offers the better potential return on margin.
Dividing the dollar edge forecasted by the scenario analysis (at our RV forecast) by the maintenance margin of each position gives:
| Position | Edge on Margin |
|---|---|
GME Short Straddle | 0.17% |
GME Iron Condor | 0.30% |
IBIT Short Straddle | 1.30% |
IBIT Iron Condor | 1.10% |
Under this account's rule-based margining, the GME iron condor is the clear preference between the two GME structures. On IBIT, the difference between the structures is small — the choice comes down to whether you're willing to delta hedge and actively manage the position, or prefer a static structure that doesn't require hedging.
I took all four positions here for illustrative purposes and retroactively walked through the analysis that should have been done up front. With the analysis done, it's clear that:
GME short straddle was not worth it on a return-on-margin or edge-on-margin basis.IBIT positions were attractive.By April 9th, the volatility conditions had shifted enough to trigger exits on both names.
GME ExitGME implied volatility had fallen below our realized-volatility forecast, while RV ticked up slightly. This combination signals exit, and we close in the market.
IBIT ExitIBIT implied volatility dropped significantly, falling below our forecast of 44.5%. The position is showing a healthy profit, and we close the spreads.
IBIT has experienced positive drift, enough that the optimal delta-hedge calculator suggests we should be buying shares to stay within our delta bands. However, since IV has dropped below our forecast, our exit condition is triggered. Rather than hedging, we close the position.
Including commissions, the results across the four positions:
| Position | Days in Trade | P&L | Return on Credit Sold |
|---|---|---|---|
IBIT Iron Condor | 14 | $283 | ~31% |
IBIT Short Straddle | 14 | $453 | ~25% |
GME Iron Condor | 14 | $199 | ~53% |
GME Short Straddle | 14 | $317 | ~40% |
Aggregate figures:
Overall these trades worked out, and we got the volatility drop we were positioned for. But they could just as easily have gone the other way. The most important thing to remember is this: a trade that loses money is not necessarily a bad trade, and a trade that makes money is not necessarily a good one. Quality of process — edge, sizing, structure selection, execution — is the only thing you control, and it's the only thing that pays off over a large enough sample.
Answer all questions correctly to complete this section.