WuBlockchain has republished a long-form article by @Boywus arguing that if traders ignore latency, infrastructure, model complexity and real-world risk constraints for a moment, market making and gamma scalping end up addressing the same blunt question: after price changes, how much should you buy, how much should you sell, and how much inventory should remain on the book?
The article says both strategies are usually wrapped in very different narratives. Market making tends to evoke ultra-fast systems, low-latency execution, sophisticated algorithms and firms such as HRT, Jump and Citadel Securities. Gamma scalping in options brings to mind pricing models, volatility surfaces, multi-leg hedges and dynamic delta management. But once everything is reduced to what the trading system actually does, the gap looks smaller. In both cases, the trader is managing inventory as a function of price.
The simplest market maker already follows an inventory curve
The piece starts with a stripped-down example of fixed-distance market making. Assume the price is 100. The trader makes no directional forecast and ignores order flow imbalance, lead-lag signals and toxic flow. They simply place a bid at 99 and an ask at 101. If the market drops to 99, they buy. If it later rises back to 101, they sell and earn 2.
That, the article says, is Fixed Distance Market Making in its most basic form. Real market making is not that crude. In live trading, participants still need to consider adverse selection, inventory pressure, order book depth, fill probability, queue position, fees and volatility. Still, if those complications are stripped out and only the final behavior is observed, the rule is straightforward: the lower the price, the more inventory the strategy accumulates; the higher the price, the less it wants to hold.
In that sense, even the simplest market-making setup already embeds an inventory curve. If price runs in one direction, the inventory becomes uncomfortable. If price oscillates, the strategy can recycle positions again and again.
Gamma scalping also turns into “sell higher, buy lower”
The article then moves to delta and gamma in options. Delta is described as the sensitivity of a derivative’s price to the underlying price. For perpetual futures, delta can be approximated as 1, which is why they are often treated as linear products. Options are different. Call delta is usually between 0 and 1, while put delta is usually between -1 and 0, and delta itself changes as the underlying moves.
Gamma is delta’s sensitivity to the underlying, or, more formally, a second-order sensitivity of the option price. The author uses a basic example. Suppose a trader buys one call with delta at 0.5 and the position is long gamma. If the trader does not want a directional bet on the underlying, they can sell 0.5 units of spot, though in practice futures or perpetuals are often used instead. The call contributes +0.5 delta and the hedge contributes -0.5 delta, leaving the book roughly delta-neutral.
That neutrality does not last because options are not linear. If the underlying rises and the call delta moves from 0.5 to 0.6, the original short hedge of 0.5 is no longer enough. The portfolio now carries +0.1 delta and the trader must sell another 0.1 to restore neutrality. If the underlying falls and call delta slips from 0.5 to 0.4, the trader has hedged too much and needs to buy back 0.1. The article describes that repeated rebalancing process as dynamic delta hedging.
For a delta-neutral long-gamma position, the behavior that finally reaches the execution layer is not mysterious at all: sell into strength, buy into weakness. As the underlying rises, call delta gets larger, so the hedge leg has to be sold more aggressively. As the underlying falls, delta gets smaller, so some of the previously sold hedge needs to be bought back. The practical questions of when to rebalance, how much to trade, and whether the hedge is continuous or discrete come later. The basic direction of trading is simple.
The author says this should sound familiar. The basic market maker behaves the same way. When price drops, bids get filled and inventory rises. When price climbs, inventory gets sold down. If the option leg is hidden and only the hedge flows are observed, long gamma plus delta hedge also becomes a process in which the trader sells more at higher prices and buys more at lower prices.
Similar execution paths do not mean identical P&L engines
The article warns against pushing the comparison too far. Fixed-distance market making and long gamma with delta hedging may look alike in execution, but they do not earn the same kind of money.
A simple market maker posts quotes at 99 and 101 and waits for others to trade against them. If they buy at 99 and sell at 101, that 2-point spread is the spread they themselves quoted. As long as the market swings back and forth, and as long as no informed trader shows up knowing the price is about to collapse to 90 and hits the 99 bid first, the market maker has a chance to keep that spread.
So the most direct revenue source for the market maker is spread capture in exchange for providing immediacy. The cost is inventory risk and adverse selection. In the article’s framing, if you stand in the market willing to take what others do not want, someone may unload the worst inventory onto you at exactly the wrong time.
Gamma scalping is different. Buying a call does not hand the trader a natural 99/101 spread. Quite the opposite. To keep re-hedging delta, the trader often needs to cross the spread and lift or hit liquidity, meaning they may be the taker paying market makers. What they truly own is a delta curve that changes with price. As price rises, the curve forces some selling. As price falls, it forces some buying. Only if actual market movement is large enough and frequent enough over time can the gains from that dynamic hedging offset the cost of buying gamma in the first place.
That is why the article separates the P&L sources so clearly. A plain-vanilla market maker sells liquidity and captures spread while taking inventory and toxic-flow risk. A long-gamma trader spends money upfront to buy convexity, then tries to realize the value of subsequent price movement through repeated delta hedging.
The article offers a practical contrast. The market maker asks whether the spread is wide enough to cover adverse selection and inventory costs. The long-gamma trader asks whether the implied volatility paid for at entry can ultimately be covered by realized volatility. The income streams differ even if the trade path can look strikingly similar.
“Synthetic gamma” without owning an option
The article then pushes the logic one step further. If the real objective is simply to obtain an inventory rule that says “buy lower, sell higher,” why must a trader start by purchasing an option? Why not write the rule directly?
The answer, according to the piece, is yes, that can be done. This is where it introduces a concept it calls “synthetic gamma” or “manual gamma.”
Imagine a trader holds no options at all and BTC is trading at 100. The trader writes a basic rule: for every 1-point decline in price, buy 1 more unit; for every 1-point increase, sell 1 unit. Inventory is no longer fixed. It becomes a function of price. The lower the price, the more the trader holds. The higher the price, the less the trader wants. In other words, the trader has manually written an inventory curve, and the target position changes each time price moves.
The article says that from the perspective of trading behavior, this starts to resemble long gamma plus delta hedge. The options trader keeps adjusting because delta changes with price. The manual trader gets to the same kind of behavior by writing a rule that directly links target inventory to price levels. The final actions can look similar even though no option is present in the account.
The author is careful to add a limit here. This does not mean a trader has created a real option out of thin air. Gamma remains the second-order sensitivity of a specific nonlinear product. But if the focus is narrowed to dynamic trading and the second-order features of realized P&L, then a trader can use linear products and discrete adjustments to reproduce part of gamma’s effect.
Grid trading is presented as the clearest example. When price drops one grid, add some exposure. When it rises one grid, cut some exposure. The strategy does not need Greek letters. It only needs a rule sheet: if price reaches 99, hold this much; if it drops to 98, hold more; if it climbs back to 100, reduce again. Every time price arrives at a new point, the system adjusts to the position specified in advance.
If those grids become tighter and tighter, not every 1% but every 0.1% or 0.01%, the resulting inventory changes start to resemble a continuous response. The article uses this to make gamma feel less abstract. A programmer can write a price-to-position rule manually through grid parameters, while an option bakes that nonlinear relationship into the product structure itself.
Fixed-distance market making fits into the same frame. On the surface, it is about posting bids below the market and offers above it to earn spread. But along the inventory path, the same mapping is still there: price determines inventory. As the market falls, the trader is passively filled and inventory rises. As it rises, inventory is sold down.
Once product labels are stripped away, the article says many strategies can be reduced to a very plain operational question: if price moves from 100 to 101, what should my position become; if it returns from 101 to 100, how much should I reverse? Gamma scalping answers through options. Market making answers through quoting rules. Grid trading answers through a schedule drawn in advance.
Nonlinearity is not free; the bill just shows up elsewhere
The article also tries to remove some of gamma’s mystique. Many traders first meet gamma through the language of second derivatives in financial engineering, which makes it sound remote. But for someone building execution systems, the author argues, a more natural starting point is this: once the position itself changes with price, the price path no longer produces only first-order P&L. It begins to generate second-order effects.
At that stage, the trader is not merely long 1 BTC and therefore up 1% when BTC rises 1%. The quantity of BTC held is also changing while the market moves.
Even so, the article says a trader cannot dodge the economic cost by skipping the option purchase and coding a “buy lower, sell higher” algorithm instead. The option merely packages the nonlinear curve and sells it directly. If a trader chooses to replicate that curve on their own, they still face trading costs, inventory risk, gap risk and discrete hedging error. In a one-way market, they can still become trapped deeper and deeper in an unfavorable position.
The point is that nonlinearity does not make cost disappear. It only moves the cost from the option premium into execution and risk management. Whether gamma is bought as a product or recreated through trading rules, it is still not free.
The “cola factory” analogy for options market making
In its closing sections, the article draws on an analogy attributed to Zhihu user “大空驴” @short_donkey. To explain option selling to non-options traders, linear assets are compared to sugar cane and options are compared to cola.
As retold in the article, sugar cane exists in nature and people have consumed it for a long time. Coca-Cola does not exist in nature. It is a processed product made from sugar cane, water, acesulfame and other inputs. Some people like sugar cane and some like cola, and the markets are not identical. But that does not mean the sugar cane business and the cola business follow the same industrial model.
Sugar cane production is not presented as a high-tech business. Its core costs may be labor and land. Cola production, by contrast, requires more than sugar cane. It involves refining, additives, preservatives, formula management, packaging design, branding and a global distribution network. The article cites the analogy to argue that cola is a higher value-added business, one that also demands larger capital commitments and stronger management. The first cup of cola, the article says, would have been expensive. Competition is what eventually turned it into a cheap mass product.
The author says the analogy is useful for beginners. Spot and futures are linear products: up 1, earn 1; down 1, lose 1. Their structure is simple, like a raw material sitting there in its natural state. Options are different. They take those linear ingredients and process them into a nonlinear payoff structure. A customer can buy a call that limits losses below a point while preserving upside, or buy a put that pays off if the market crashes. In both cases, what is being purchased is a redesigned distribution of outcomes.
Those nonlinear payoffs do not descend from the sky. When a client buys an option, all the client sees is an extra call or put in the account. The market maker on the other side has to manufacture that product. According to the article, that can mean combining spot, futures, options at other strikes and options with other maturities, then continuously running delta hedges while managing gamma, vega, inventory and tail risk.
The customer sees a can of cola with a label already attached. The market maker sees sugar, water, bottles, machines, logistics and the production line. From this angle, selling options resembles a processing industry. The raw materials are basic first-order assets from the market. Models, hedging, execution and capital are used to process them into the nonlinear payoff the client wants, which is then offered at a price.
The article says that a professional options market maker is not just guessing whether the market will rise or fall tomorrow, and not merely guessing future volatility. The real business question is whether the “factory” can produce the bottle of cola the client wants at a cost lower than the price the client is willing to pay.
That, in the author’s view, helps explain why two market makers can get very different outcomes while quoting the same option into the same market. Some can run delta hedges cheaply, with lower fees, smaller slippage and internal order flow that offsets risk efficiently. Others have slower systems, worse fee terms or poor inventory management, and end up chasing the book each time the market moves. The visible selling price may be the same, but each factory’s production cost is different.
The article extends the point to options pricing more broadly. Pricing models are teachable, and Black-Scholes is not proprietary knowledge. The real edge comes from turning a theoretical value into a tradable real-world price and then disposing of customer risk cheaply once it arrives.
From two buttons to an industrial process
The piece closes by reconnecting all of this to market making. The simplest market maker posts two-sided quotes and earns a few ticks of spread. More advanced firms add inventory control, dynamic pricing and order flow prediction. In options market making, “inventory” no longer means only a spot position. It expands into a full set of delta, gamma, vega and term-structure exposures.
Still, the article argues that the deeper questions remain old ones. What risk have I taken on? How do I hedge it? What does the hedge cost? And does the money the client pays cover that cost?
That is where the article sees the most interesting connection between gamma trading and market making. At the beginning, a trading system has only two buttons: buy and sell. Later, traders attach inventory rules to those buttons. Later still, they describe those rules with delta and gamma. In the final stage, someone builds a factory and packages those rules into products that can be sold to other people.
That, the author says, is the path from the dumbest possible market maker to selling cola.