Side Lesson 20: Greeks Deep Dive — Vanna, Charm, Color, and Second-Order Surprises

---

1. Why This Is Important

Week 29 introduced the five first-order Greeks — delta, gamma, theta, vega, and rho — as the partial derivatives of an option's price with respect to the inputs that matter (spot, time, volatility, rates). For 95% of retail option positions, those five are everything you need. Size your covered call by delta, watch your theta, glance at vega before earnings, and you are done.

But options live on a non-linear surface, and the first-order numbers are only the local slope. Once a portfolio is large enough — or once you start hedging dynamically, or running 0DTE flow — the second-order Greeks start to matter:

Delta itself moves when volatility moves (vanna). This is why a

sharp VIX spike during a sell-off does not hurt your puts as much as you would have predicted from delta alone.

**Delta drifts as time passes even when the stock does not move

(charm).** Your weekend hedges are wrong by Monday open. ATM options' delta plus charm explains why pin risk is real.

Gamma changes as expiration approaches (color). That smooth

gamma profile from week 29 turns into a needle on expiration day. Color is what describes the steepening.

Vega itself has convexity (volga / vomma). A 1-point move in

IV does not change vega linearly. Volga is why deep OTM "lottery tickets" can quintuple in a single VIX-66 day.

There is a fifth, broader reason this lesson exists: **dealer positioning has become a first-order driver of intraday SPX behavior since 2022.** The explosion of zero-DTE (0DTE) flow on the index options has flipped the sign on aggregate market-maker gamma multiple times per week. June and December quarterly OpEx weeks now show clear pin patterns where SPX hugs a round-number strike like a magnet into the 4 PM print. None of that flow is visible if you only know delta. Knowing where the second-order Greeks are concentrated is how you read it.

The honest framing for retail, though — and this lesson hammers it in three places — is that **unless you have more than ~5% of your net worth in options, second-order Greeks are interesting, not actionable.** Read this lesson the way you would read a book on engine internals: you will be a better driver, but you do not have to rebuild your carburetor.

---

2. What You Need to Know

2.1 First-Order Greeks — One-Line Refresher

For a European call priced via Black-Scholes with spot $S$, strike $K$, time-to-expiry $T$, volatility $\sigma$, and risk-free rate $r$:

• Delta $\Delta = \partial C / \partial S = \Phi(d_1)$ — the

hedge ratio. 0 to 1 for calls, 0 to -1 for puts. Quoted per $1 of spot move.

• Gamma $\Gamma = \partial^2 C / \partial S^2 = \phi(d_1) /

(S \sigma \sqrt{T})$ — the curvature. Always positive for long options. Bell-shaped, peaks at-the-money.

• Theta $\Theta = \partial C / \partial t$ — the daily decay,

negative for long options. Most negative at-the-money, accelerates into expiry as $1/\sqrt{T}$.

• Vega $\nu = \partial C / \partial \sigma = S \phi(d_1) \sqrt{T}$

— the IV sensitivity. Quoted per 1 vol-point. Peaks at-the-money, rises with $\sqrt{T}$.

• Rho $\rho = \partial C / \partial r = K T e^{-rT} \Phi(d_2)$ —

the rate sensitivity. Negligible for short-dated options, matters for LEAPS (week 38).

That is the table you can read off any broker platform. Everything below this section is the derivatives of those derivatives.

2.2 Vanna — When IV Moves, Delta Moves

Vanna is the cross-derivative

$$ \text{vanna} = \frac{\partial \Delta}{\partial \sigma} = \frac{\partial \nu}{\partial S} = -\phi(d_1) \frac{d_2}{\sigma} $$

In words: how much your delta changes per 1 vol-point change in IV.

The classic case: you are long a -25-delta SPY put as a tail hedge. The market sells off 4%, but at the same time VIX jumps from 14 to

Your put's delta has shifted from -0.25 to roughly -0.42

even before you account for the spot move, because vanna for a put of that strike is positive (-d2 is positive when the put is OTM, and the negative sign in the formula flips). The hedge works *better than the static delta would suggest* — that is the vol-tail-wags-dog mechanic, expressed in a Greek.

Same mechanic, opposite direction: a covered-call seller above the money sees their short-call delta rise when IV pops, meaning their effective long stock exposure shrinks faster than expected. That is a vanna assignment risk.

Where vanna concentrates: OTM options, both calls and puts. At-the-money, $d_2$ is near zero, so vanna is near zero too. The sign rules of thumb (call vanna > 0 for OTM, < 0 for ITM; puts mirror) come straight from the sign of $d_2$.

The top-right panel of the image above shows vanna's S-curve: peaks of opposite sign on either side of the money, zero at ATM. This is why dealers obsessing over their book's vanna exposure focus their attention on the 25-delta and 10-delta wings, not on the 50-delta strike.

2.3 Charm — When Time Passes, Delta Moves

Charm (also called delta decay) is

$$ \text{charm} = \frac{\partial \Delta}{\partial t} = -\frac{\partial \Delta}{\partial T} $$

For a call with $r=q=0$ this simplifies to a clean

$$ \text{charm}_{\text{call}} = -\phi(d_1) \cdot \frac{d_2}{2 T} $$

What that means in trading terms: **even if SPY does not move at all, your delta drifts overnight.** A 60-day, -30 delta put on SPY with 30% IV picks up about +0.0035 of delta per day from charm alone. Stack a Friday-to-Monday weekend in there and the position has shifted ~0.01 of delta before the open Monday — just from time elapsing.

Practical implications:

Weekend gap risk is partly charm risk. If you delta-hedged

Friday at 4 PM, you will be off-hedge on Monday at 9:30 AM by roughly $\text{charm} \times 3$ days, even if futures opened unchanged. Delta-neutral books rebalance for charm into Friday's close.

Pin risk on expiration day is charm pushed to the limit. ATM

options with hours to expiry have $|d_2|$ close to zero but $T$ in the denominator collapsing — charm explodes. That is what makes it impossible to know whether you are going to be assigned on a Friday $K = S$ short call until the print.

Calendar spreads have charm exposure by construction. The

short leg's delta drifts faster than the long leg's. If you buy a calendar at zero net delta, you are long charm — your delta will swing one way or the other as time passes.

2.4 Color — When Time Passes, Gamma Moves

Color (or gamma decay) is

$$ \text{color} = \frac{\partial \Gamma}{\partial t} $$

Where charm describes how delta drifts with time, color describes how gamma drifts with time. This is the Greek that gets quoted in dealer-positioning reports under names like "gamma roll-off" or "gamma re-loading."

The shape: gamma is a bell. As $T \to 0$, the bell **gets taller and narrower**. Color is the rate at which that narrowing happens. The center of the bell (ATM) sees gamma rise toward infinity in the last hours; the wings see gamma collapse to zero.

A trading-floor way to phrase it: *"this morning my book is short $5M of gamma at 4980 strike; if SPX stays here through close, color makes it $7M short by Wednesday."* The position has not moved, the underlying has not moved, but the risk has grown.

2.5 Volga (Vomma) — Vega's Own Convexity

Volga is the second derivative of price with respect to vol:

$$ \text{volga} = \frac{\partial \nu}{\partial \sigma} = \nu \cdot \frac{d_1 d_2}{\sigma} $$

It tells you whether your vega exposure itself accelerates or decelerates as IV moves. The sign is positive when $d_1 d_2 > 0$, which means deep OTM and deep ITM options have positive volga while ATM options have volga near zero (because $d_2 \approx 0$ there).

Why this matters: "lottery ticket" deep-OTM puts during a vol spike. A 10-delta SPY put bought at IV 18% is worth $0.40. The market sells off 6%, IV pumps to 38%, and the put — even ignoring the move in spot — multiplies because vega itself rose as the strike moved closer to the money in vol-distance terms. Volga is why short-vol traders famously blow up on tail events: their vega got shorter faster than they could hedge it.

2.6 Dealer Positioning, Gamma Walls, and 0DTE

This subsection is the one that has changed the most since 2022.

Pre-2022: the bulk of SPX option open interest was monthly expirations, with quarterly OpEx (the third Friday of March, June, September, December) being the largest. Dealers' aggregate gamma sat predominantly at round-number strikes (5000, 5100, 4900). On expiration week — Wednesday especially, when index gamma starts to roll off — SPX often "pinned" near a major strike because the dealers who were short gamma at that strike had to keep buying the dip and selling the rip to stay delta-neutral.

Post-2022: the 0DTE explosion. Daily expiries on SPX/SPY went from a niche product to ~45% of total SPX option volume by mid-2024. Dealers' net gamma can flip from long to short multiple times per day. The pinning patterns are now both more frequent (because expiry is every day) and more violent (because charm and color are extreme on a same-day option).

The image above is an illustrative reconstruction of an SPX cash session on a recent quarterly OpEx Friday, with the index drifting through the morning, snapping toward 5000 by lunch, and effectively oscillating in a 0.15% band around that level for the final two hours. The dotted horizontal lines mark the three highest-OI strikes into the print. None of that is mechanical destiny — but it is now common enough that desk research at every prime broker tracks it.

The retail takeaway: if you are short an iron condor whose short leg is a major round-number strike during quarterly OpEx week, **do not chase price action** if SPX seems "magnetized" to that strike on Wednesday-Thursday-Friday. The behavior is explained, not magic; it is also reversible the moment options expire and dealers re-hedge. Your edge is patience.

2.7 The Retail Filter — When Second-Order Greeks Actually Matter

Here is the rule, repeated because it is important:

**Unless options are more than ~5% of your investable net worth, do not optimize for second-order Greeks.** The first-order ones, plus basic position sizing, will dominate any difference. The marginal edge from understanding charm is real but small for someone with a single covered-call program or a quarterly tail hedge.

The exceptions where it does start to matter:

You run a delta-hedged position. The moment you start

re-hedging dynamically, you are making the second-order Greeks into your P&L drivers. Vanna and charm move your delta between re-hedges; volga and color move your gamma exposure.

You hold options through the last week of expiry. Charm,

color, and theta acceleration interact non-trivially. If you are still in the position at T<7d, you need the full picture.

You write spreads against earnings. Volga is the Greek that

blows up short-strangles when "implied move" is breached.

You concentrate in 0DTE. Every Greek in this lesson is

first-order on a same-day option.

Outside of those four cases, the lesson lives in your peripheral vision: useful context, not a hedging instruction.

---

3. Common Misconceptions

"Higher-order Greeks are just academic." They were a footnote

pre-2022. Post-2022, with 0DTE and large quarterly OpEx, vanna and charm flows are observable in intraday SPX data. They are academic only if you ignore the index that 70% of US retail equity is benchmarked to.

"Vanna is always positive for long options." Vanna's sign

depends on moneyness. Long OTM calls have positive vanna; long ITM calls have negative. The mistake comes from confusing vanna (cross-derivative) with vega (which is always positive for long options).

"Charm is just theta." No — theta is the decay of the *option

price; charm is the decay of delta*. A position can have negative theta and positive charm simultaneously (a long OTM call burns premium daily but its delta drifts toward zero in absolute value on the same time axis).

"Gamma is constant on expiration day." It is the opposite —

gamma is least stable on expiration day. Color (the rate at which gamma changes per unit time) is largest near expiry, and gamma itself can spike toward infinity at exactly $S = K$, $T = 0$.

"Dealers always pin SPX to round strikes." The pinning

mechanic requires dealers to be net-short gamma at the strike. If they are net-long gamma (because customer flow has been buying options instead of selling), the same setup produces anti-pinning — magnification of moves, not damping.

"0DTE is just a faster casino." The 0DTE flow has changed the

structure of intraday SPX volatility. Vanna and charm flows on 0DTE can produce identifiable intraday patterns (morning chop, afternoon trend) that did not exist five years ago. It is a real structural change, not a fad.

"Volga only matters during VIX spikes." Volga matters during

any large vol move. The asymmetry — vol can rise faster than it falls — is what creates a positive expected return for being long volga in a tail-hedge construct.

"Once you understand the formulas, you can hedge them out."

Hedging vanna requires options at different strikes; hedging color requires options at different expiries. Each hedge adds transaction cost, and each adds residual risk in the Greek you used as a hedging instrument. Institutional desks accept residuals in third-order Greeks.

"Higher-order Greeks always make positions worse." Long-vega

structures often have positive volga, which is convexity working for you during a vol expansion. The question is what you paid for it (theta) versus how much vol-of-vol you actually got.

"All these Greeks scale linearly with size." They scale

linearly per contract, but the underlying spot and IV processes do not scale linearly. A 100-contract position behaves differently from 10 ten-contract positions across different strikes — diversification across the smile changes the aggregate Greek profile.

---

4. Q&A Section

**Q1: If charm changes my delta overnight, why doesn't my broker re-hedge for me?**

It does not, because it does not know your hedging policy. Brokers report your portfolio Greeks (delta, gamma, theta, vega) but do not auto-rebalance hedges. If you run a delta-neutral book, you place the rebalance order yourself based on the morning's portfolio delta. Charm is one of the inputs your hedging algorithm should account for; weekend charm is the most common one missed.

Q2: Can I observe vanna flows on a public chart somewhere?

Indirectly. Several research desks (SqueezeMetrics, SpotGamma, Cboe research) publish daily estimates of dealer gamma exposure ("GEX") and vanna exposure. The numbers are estimates because dealer books are not public. The quality is "directionally useful" not "tradeable on its own." Treat them like sentiment indicators.

Q3: What's the simplest strategy whose primary Greek is volga?

A short ATM straddle is short volga (you lose convex amounts when IV expands). Conversely, a long strangle, especially deep OTM, is long volga. The volatility risk premium (week 49) is essentially the market paying a premium to insure the negative-volga seller against tail-vol expansions.

Q4: Why do option books often "pin" near round strikes specifically?

Two reasons. First, customer flow tends to cluster at psychologically round strikes (5000, 5100), so open interest is highest there. Second, those round strikes attract index-arb hedging from ETF market-makers and SPX/SPY converters. The combined effect is more gamma sitting at the round number than at, say, 4983, which means dealers' gamma-induced rebalancing pressure peaks there.

Q5: Does color really matter to a covered-call seller?

Marginally. If you write a 30-day covered call, color tells you that the position's gamma profile narrows over the 30 days — by day 25, your call's gamma is concentrated in a narrow band around the strike. This affects assignment probability if the stock starts the final week near the strike. Most retail covered-call writers do not care, but it is the technical reason "rolling at 21 DTE" is the common rule: that is the date by which the gamma re-concentration becomes meaningful.

Q6: Is vanna positive or negative for puts?

Same formula, opposite sign convention. For a put, vanna $= -\phi(d_1) \cdot d_2 / \sigma$ as well. The interpretation flips: when you are long an OTM put and IV rises, your delta becomes more negative (closer to -0.5 from -0.3), which actually makes you *more bearish- hedged*. That is the structural reason vol spikes during equity sell-offs make tail hedges more effective than they "should" be.

Q7: Are there third-order Greeks?

Yes — speed (∂Γ/∂S), zomma (∂Γ/∂σ), ultima (∂Volga/∂σ). They are relevant to specialist desks running large variance-swap books or exotic structures. For US-listed vanilla options, third-order Greeks are noise-level.

Q8: How do retail platforms display these?

Inconsistently. ThinkorSwim and Interactive Brokers expose Vanna, Charm, and Volga in their analytics tabs but bury them. Fidelity, Schwab, Robinhood do not display them at all. If you want them, you compute them yourself from the BSM closed-form (the interactive on this page does it live).

Q9: How does this connect to week 40 and week 49?

Week 40 (VIX) explained that VIX itself is the expected variance of SPX. Week 49 (vol arbitrage) explained the volatility risk premium. **The second-order Greeks are how those macro vol stories land on individual option positions.** The dealer flows that drive VIX and the VRP are the same flows you're seeing in vanna / charm / color exposure. Same physics, different aggregation level.

Q10: Should I add a "vanna trade" to my portfolio?

Almost certainly not. Vanna trades — long/short combinations designed to isolate vanna exposure — require liquid options across the surface, low transaction costs, and real-time risk management. For retail, the pure-second-order trade is a dead-end. Use second- order Greeks as explanation tools for what your existing positions are doing, not as a strategy menu.

Q11: What's the underlying lesson here?

Vol-tail-wags-dog. Vanna and charm are the mechanism by which vol expansions and time decay rearrange equity exposure without anyone trading the underlying. Barbell sizing and options-tax discipline matter too, but the central idea is the vol tail.

Q12: If I only remember one thing from this lesson?

"Greeks beyond the first five matter when you are running options as a system, not as a position." A single tail hedge once a quarter? First-order is enough. A daily-rebalanced 0DTE book? You need the whole stack — and you should not be running that book alone.