WhitmanTrading

Colour: How Fast Gamma Itself Decays

Colour is a third-order option greek: the rate at which gamma changes as time passes. It tells a hedging desk how quickly today's hedge will go stale, and it grows sharply as expiry approaches. Almost nobody outside a desk has a use for it.

Defined exactly

A candlestick chart of the site's shared price history. The headline on the chart reads: How fast gamma itself decays with time.
How fast gamma itself decays with time. Illustrative chart - not real market data.

Colour is the rate at which gamma changes as time passes. Delta measures how the option price responds to the underlying. Gamma measures how delta responds. Colour measures how gamma responds to the clock — one more step down the same chain.

A gently rising stretch of the long price series with an account equity curve beneath it. The headline on the chart reads: It is a third-order greek, and that is the whole story.
It is a third-order greek, and that is the whole story. Illustrative chart - not real market data.

Third-order is the important word. Each order of differentiation moves further from anything you can observe and closer to something the model asserts. Delta you can almost feel on a position. Gamma you can notice over a day. Colour you can only compute.

A calmly advancing stretch of the long price series with a slowly rising equity curve beneath it. The headline on the chart reads: It tells a hedger how stale today's hedge will be.
It tells a hedger how stale today's hedge will be. Illustrative chart - not real market data.

It has exactly one job. A desk that has hedged a book to delta-neutral wants to know how long that neutrality lasts. Gamma says how much the hedge drifts when the underlying moves; colour says how much that sensitivity itself changes overnight, which is how you schedule the next rehedge.

A flat, quiet stretch of the long price series with a gradually rising equity curve beneath it. The headline on the chart reads: And it grows sharply as expiry approaches.
And it grows sharply as expiry approaches. Illustrative chart - not real market data.

Near expiry the number becomes large and unstable. Gamma on a near-the-money option close to expiry changes violently from one hour to the next, and colour is the measurement of that violence. It is the formal statement of something every options trader learns the hard way — that the last few days behave nothing like the preceding weeks.

Who this is actually for

A strongly rising stretch of the long price series with an account curve breaching its limit. The headline on the chart reads: Which matters to a desk and almost nobody else.
Which matters to a desk and almost nobody else. Illustrative chart - not real market data.

It is a market-making tool. A desk holding thousands of positions and rehedging many times a day needs to know how its risk profile decays between rehedges, because the cost of getting that wrong is paid continuously across an enormous book.

A choppy, directionless stretch of the long price series. The headline on the chart reads: A retail position is not hedged often enough to care.
A retail position is not hedged often enough to care. Illustrative chart - not real market data.

If you rehedge rarely or never, the number tells you nothing you can use. Colour is about the interval between adjustments. A position adjusted twice in its life has no interval worth optimising, and the decision that mattered was the size at entry.

A declining stretch of the long price series. The headline on the chart reads: And it is a model output, not a market price.
And it is a model output, not a market price. Illustrative chart - not real market data.

Nobody quotes colour and no exchange publishes it. It falls out of whichever pricing model you use, and a different model with different volatility assumptions produces a different number for the same position. Two of the inputs behind it are assumptions rather than observations, which is worth remembering before treating the output as a measurement.

In practice

A 72-bar candlestick section of the shared price history with an account curve shown with and without fees. The headline on the chart reads: Rehedging is what costs money, not the greek.
Rehedging is what costs money, not the greek. Illustrative chart - not real market data.

Every rehedge is a real transaction with a real cost. The greek is free; acting on it is not, and the whole point of computing it is to rehedge as infrequently as the risk allows rather than as often as possible.

A candlestick chart with a volume histogram beneath it, with the volume histogram emphasised. The headline on the chart reads: Participation does not enter the calculation.
Participation does not enter the calculation. Illustrative chart - not real market data.

No volume appears anywhere in it. The inputs are price, strike, time, rate and volatility. Who is trading and how much is not part of the arithmetic, even though it is very much part of whether you can execute the hedge.

A long-horizon candlestick view of the same price series. The headline on the chart reads: On a long-dated option it is nearly nothing.
On a long-dated option it is nearly nothing. Illustrative chart - not real market data.

With months to run, the number is negligible. Gamma changes slowly when there is plenty of time left, so the rate of that change is close to zero and the hedge holds for days.

A candlestick series containing several opening gaps, with the largest opening gap marked. The headline on the chart reads: And a gap makes every greek stale at once.
And a gap makes every greek stale at once. Illustrative chart - not real market data.

An opening gap invalidates the whole set simultaneously. The greeks describe smooth, small movements. A jump is neither, and the careful overnight calculation of how the hedge would drift is simply overtaken by an event it does not model.

A declining stretch of the long price series, with the entry price and the level at which a stop would trigger drawn as horizontal lines. The headline on the chart reads: No greek tells you where the stop belongs.
No greek tells you where the stop belongs. Illustrative chart - not real market data.

None of the greeks contain a risk rule. They describe sensitivities. Where a stop goes, and therefore what size is defensible, is a separate decision that no amount of differentiation will make for you.

A candlestick chart of the site's shared price history, annotated with the round-trip cost. The headline on the chart reads: Every rehedge costs 2% of a bar.
Every rehedge costs 2% of a bar. Illustrative chart - not real market data.

On this site’s shared price history a round trip is 2% of a median bar’s range. Multiply that by a hedge adjusted several times a day and the cost of precision quickly exceeds the risk it was removing.

The useful version of this idea

Hedges have a shelf life, and it shortens as expiry approaches. That sentence is the whole practical content, and you can act on it without computing anything.

Which means the decision to hold an option into its final days is a decision to accept faster-changing risk. Positions that behaved predictably for weeks start moving differently every few hours, and any plan that assumed the earlier behaviour is now wrong. Closing or rolling before that period is a legitimate choice made in advance, rather than a reaction to a position that suddenly stopped cooperating.

And the second practical use is scepticism. When a platform displays a long list of greeks to three decimal places, the precision is real and the accuracy is not — every one of those numbers rests on a volatility assumption that nobody can observe.

What colour is not

It is not a market price. No one quotes it.

It is not a trading signal. It describes decay.

It is not needed by most traders. Rehedging is rare.

And it is not a risk rule. It contains no stop.

When it fails

A sideways, range-bound candlestick series. The headline on the chart reads: In a quiet market it is the calendar doing the work.
In a quiet market it is the calendar doing the work. Illustrative chart - not real market data.

In a still market the number keeps changing anyway, because time keeps passing. That is the honest description of what it measures and also the reason it is easy to misread: something is moving on the screen while nothing is happening in the market, and the movement is the clock rather than information.

The second failure is treating a model output as a measurement. Change the volatility input and every greek changes with it.

A third is using it at all as a retail trader. The rehedging interval it optimises does not exist for most positions.

A fourth is trusting any greek through a gap. They describe small smooth moves and nothing else.

A fifth is confusing precision with accuracy. Three decimal places on an assumption is still an assumption.

And a sixth is rehedging more often because the number is large. The transaction cost is certain and the risk reduction is not.

A worked example

Take a position that is delta-neutral at the close, with three weeks to expiry. Gamma is moderate, so overnight movement in the underlying will change the delta somewhat, and colour is small, so gamma itself will be about the same tomorrow as it is today. The hedge is roughly still a hedge in the morning.

Now take the same position with three days to expiry. Gamma is far larger and colour is larger still. The overnight change in delta is bigger, and the sensitivity producing that change is itself different by morning. The hedge set at the close is not the hedge you would set at the open, and the gap between them is what a desk is paid to manage and what an individual trader is usually better off avoiding entirely.

The practical conclusion is not a formula. It is that the same position, the same size and the same strike require completely different attention depending only on how many days remain — and that the change is not gradual.

The original data

Of the 24,971 videos in research/search-study-corpus.jsonl, 886 have “options” in the title, at a median of 10,455 views across 495 channels. That makes options one of the most heavily covered subjects in the corpus. The counts are in research/corpus-coverage.json, produced by site/measure_corpus.py.

A strongly rising stretch of the long price series, cut short at the decision bar. The headline on the chart reads: Gamma is rising fast into expiry. Rehedge?
Gamma is rising fast into expiry. Rehedge? Illustrative chart - not real market data.

886 videos on options, and the third-order greeks appear in essentially none of them. That is a reasonable allocation of attention rather than a gap in the coverage — this is a desk instrument, and a page arguing that most readers do not need it is more honest than one implying they do.

The answer to that final question depends entirely on whether you are running a book. A desk rehedges, because the cost is spread across thousands of positions and the risk is real. An individual holding a few contracts should usually close or roll instead — the transaction cost of chasing neutrality on a small position reliably exceeds the risk being neutralised.

Gamma is the second-order greek this one measures the decay of, and the one worth understanding first. Theta is the other time-based greek and the one that actually affects a held position’s value. And options covers the instrument all of these describe.

What I actually do

The useful thing I took from the third-order greeks is not the greeks. It is the realisation that a hedge has a shelf life, and that the shelf life shortens as expiry approaches. You do not need the formula to act on that. You need to know that a position which was neutral this morning may not be neutral this afternoon, and that the closer expiry gets, the faster that happens.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.