Author:AzumaOdaily Planet Daily
In the past two days, there has been a lot of discussion on X about the formula for prediction markets: Yes + No = 1. The origin of this discussion can be traced back to a post by the big name DFarm (@DFarm_club), who wrote an article analyzing Polymarket's shared order book mechanism. This sparked a collective emotional resonance with the power of mathematics — Original article link:A Comprehensive Explanation of Polymarket: Why YES + NO Must Equal 1?》，strongly recommend reading the original.
In the related discussions, several prominent figures, including Blue Fox (@lanhubiji), mentioned that "Yes + No = 1" is another minimalist yet powerful formula innovation following "x * y = k," with the potential to unlock a trillion-dollar information flow trading market. I completely agree with this view, but at the same time, I also feel that some of the discussions are overly optimistic.
The key lies in the issue of liquidity creation. Many people might think that "Yes + No = 1" solves the barrier for ordinary users to participate in market making, so the liquidity in prediction markets would naturally increase, just like in AMMs where x * y = k leads to rising liquidity. However, the reality is far from that.
The difficulty of market making in prediction markets is inherently higher.
In a practical environment, whether one can enter the market to provide liquidity and build market depth is not merely a question of participation thresholds, but more fundamentally an economic question of whether it is profitable. When compared horizontally with AMM markets based on the x * y = k formula, the difficulty of market making in prediction markets is actually significantly higher.
For example, in a classic AMM market that strictly follows the formula x * y = k (such as Uniswap V2), if I want to provide liquidity for the ETH/USDC trading pair, I need to deposit ETH and USDC into the pool in a specific ratio according to the real-time price relationship of the two assets in the liquidity pool. After that, as the price relationship fluctuates, the amounts of ETH and USDC I can retrieve will vary accordingly (i.e., the well-known impermanent loss). At the same time, I can also earn trading fees. Of course, the industry has made many innovations around the basic x * y = k formula afterward. For example, Uniswap V3 allows liquidity providers to concentrate their liquidity within a specified price range to pursue a higher risk-return ratio, but the fundamental model remains unchanged.
In this market-making model, if trading fees within a certain time period can cover the impermanent loss (and often it takes even longer to accumulate these fees), it can be profitable — as long as the price range isn't too aggressive, I can be quite lazy about market-making, just checking in occasionally every few days. However, in prediction markets, if you adopt a similar attitude toward market-making, you will most likely end up losing everything.
Let's take another example with Polymarket. Suppose we first create a basic binary market, for instance, a market where the question is "YES, the real-time price is $0.58." If I want to provide liquidity in this market, I can place a YES buy order at $0.56 and a YES sell order at $0.60. As explained in the DFarm article, this is essentially equivalent to placing a NO buy order at $0.40 and a NO sell order at $0.44. In other words, by using the market price as a reference, we provide order support at specific points slightly above and below the current price.
Now that the order is placed, can I just lie back and ignore it? Well, by the time I next check, I might see one of the following four situations:
Both two-way orders have not been executed;
Both two-way orders have been executed;
An order in a certain direction has been executed, and the market price remains within the original order range;
An order in a certain direction has been executed, but the market price has moved further away from the remaining limit orders. For example, YES was bought at 0.56, a sell order remains at 0.6, but the market price has dropped to 0.5.
So, under what circumstances can you make money? What I can tell you is that in low-frequency attempts, different situations may lead to different profit or loss outcomes. However, if you continue to operate in such a passive manner over the long term in a real-world environment, the final result will almost certainly be a loss. Why is this the case?
The reason lies in the fact that predicting markets inherently follow an order-book market-making model similar to CEXs, rather than the liquidity pool market-making logic of AMMs. The operational mechanisms, operational requirements, and risk-return structures of these two models are entirely different.
In terms of operational mechanisms, AMM (Automated Market Maker) market making involves depositing capital into a liquidity pool to collectively provide liquidity. The liquidity pool spreads liquidity across different price ranges based on the formula x * y = k and its variations. In contrast, order book market making requires placing buy and sell orders at specific price levels. Liquidity is only available where orders are placed, and trades must be executed through the matching of these orders.
In terms of operational requirements, AMM market making only requires depositing both-side tokens into the pool within a specific price range, and as long as the price does not move outside this range, the liquidity provision remains effective. In contrast, order book market making requires active and continuous order management, constantly adjusting quotes to respond to market changes.
In terms of risk-return composition, AMM market making mainly faces the risk of impermanent loss, with profits coming from transaction fees in liquidity pools. Order-book market making, on the other hand, must deal with inventory risk in one-sided market conditions, with returns derived from bid-ask spreads and platform subsidies.
Continuing with the hypothetical case from the previous example, it is known that the main risk I face when providing liquidity on Polymarket is inventory risk, while the main sources of profit are the bid-ask spread and liquidity subsidies from the platform (Polymarket provides liquidity subsidies for certain orders that are close to the market price in some markets; for more details, please refer to the official homepage). The potential profit and loss scenarios for the four cases are as follows:
First scenario: unable to capture the bid-ask spread, but can capture liquidity subsidies;
Second, profits have been made through the bid-ask spread, but liquidity subsidies will no longer be received;
Third, in the third scenario, a YES or NO position has already been taken, turning into a directional exposure (i.e., inventory risk), but in some cases, it can still capture a certain amount of liquidity premium.
In the fourth scenario, the position has also become directional, resulting in a floating loss, and there is no longer access to liquidity premiums.
There are two more points to note here. First, the second scenario is actually always evolved from the third or fourth scenario, because usually only one side of the limit orders will be executed first, which can temporarily result in directional positions. However, the risk ultimately does not materialize, as the market price later moves in the opposite direction and triggers the other side's limit order. Second, compared to the relatively limited market-making profits (which are typically fixed in terms of spread gains and subsidies), the risk of directional positions is often unlimited (the maximum risk is that all of your YES or NO tokens could become worthless).
In summary, if I want to consistently make profits as a market maker, I need to seize profit opportunities as much as possible while avoiding inventory risk. Therefore, I must actively optimize my strategies to maintain the first scenario as much as possible, or quickly adjust the order range after one side of the orders is triggered, trying to convert it into the second scenario. This helps avoid staying in the third or fourth scenarios for a long time.
It is not easy to maintain this over the long term. Market makers first need to understand the structural differences across various markets, comparing aspects such as subsidy levels, volatility, settlement times, and judgment rules. Then, they must accurately and quickly track—or even predict—price changes based on external events and internal capital flows. Subsequently, they must promptly adjust their orders proactively in response to these changes, while also pre-designing and managing execution strategies to address inventory risks. Clearly, this goes far beyond the capabilities of an average user.
A wilder, more volatile, and less rule-following market
At first glance, this seems acceptable. After all, order book mechanisms are not new; they remain the dominant market-making mechanism on CEXs and perpetual DEXs. Market makers active in these markets can easily transfer their strategies to prediction markets to continue generating profits while providing liquidity to the latter. However, the reality is not that simple.
Let's think about this question: what is the worst scenario for market makers? The answer is simple—unidirectional price trends. This is because unidirectional trends often continuously amplify inventory risks, eventually breaking through risk configurations and causing massive losses.
However, compared to traditional cryptocurrency trading markets, prediction markets are inherently wilder, more volatile, and less bound by conventional ethics, where one-sided market movements tend to be more extreme, abrupt, and frequent.
"More wild" means that in the traditional cryptocurrency trading market, if we extend the timeline, major assets will still exhibit a certain degree of fluctuation, with rising and falling trends typically shifting in cycles. However, in prediction markets, the traded assets are essentially event contracts, each with a clearly defined settlement time. The formula Yes + No = 1 determines that ultimately, only one contract's value will become $1, while the other options will all become zero. This means that wagers in prediction markets will eventually conclude in a one-sided market outcome from a specific point in time. Therefore, market makers need to design and implement inventory risk management more strictly.
"More jumps" means that in conventional trading markets, volatility is determined by the ongoing interplay between sentiment and capital. Even if the fluctuations are intense, the price changes remain continuous, allowing market makers time to adjust their inventory, control spreads, and perform dynamic hedging. However, in predictive markets, volatility is often driven by discrete real-world events, and price changes tend to be abrupt— for example, the price might be at 0.5 one second, and a real-time event can push it directly to 0.1 or 0.9 the next second. Moreover, it is often difficult to predict exactly when or which event will cause a dramatic change in the price, leaving market makers with very little reaction time.
The phrase "more unscrupulous in market ethics" means that in the market, there are many insider players who are either close to the source of information or are the information sources themselves. These players do not engage in trading based on predictions of future market conditions, but rather enter the market with clear knowledge of the outcomes in order to profit from them. In the face of these players, market makers are naturally at a disadvantage in terms of information, and the liquidity they provide becomes a channel for these insiders to cash out. You might ask, don't market makers also have inside information? This is also a typical paradox: if I already know the inside information, why would I bother making markets? I would be better off directly betting on the direction of the market to make even more money.
It is precisely because of these characteristics that I have long agreed with the saying that "the design of prediction markets is structurally unfriendly to market makers." I also strongly advise ordinary users against easily getting involved in market making.
Does this mean making markets in prediction markets is unprofitable? Not necessarily. Luke (@DeFiGuyLuke), the founder of Buzzing, once disclosed that based on market experience, a relatively conservative estimate is that market makers on Polymarket can earn about 0.2% of the trading volume as profit.
In short, this is not a way to easily make money. Only professional players who can accurately track market changes, promptly adjust their orders, and effectively implement risk management can sustain operations over a long time period and earn money through their genuine skills.
Predicting market trends may not lead to a flourishing diversity of outcomes.
The challenge of market-making in forecasting markets not only imposes higher requirements on the capabilities of market makers, but also presents a challenge for platforms in building liquidity.
The difficulty of market making implies constraints on liquidity provision, which directly affects users' trading experience. To address this issue, leading platforms like Polymarket and Kalshi have chosen to offer real financial incentives to subsidize liquidity, thereby attracting more market makers to participate.
Nick Ruzicka, an analyst specializing in prediction markets, cited a Delphi Digital report in November 2025, stating that Polymarket had invested approximately $10 million in liquidity subsidies. At its peak, the platform was spending over $50,000 per day to attract liquidity. However, as Polymarket solidified its leading position and brand influence, it significantly reduced its subsidy efforts. On average, it now still needs to provide a subsidy of $0.025 for every $100 of trading volume.
Kalshi also has a similar liquidity incentive program and has already allocated at least $9 million for this purpose. In addition, in 2024, Kalshi leveraged its compliance advantages (Odaily note: Kalshi is the first prediction market platform to receive regulatory approval from the CFTC; Polymarket also received approval in November 2025) to sign a market-making agreement with Susquehanna International Group (SIG), one of Wall Street's top market-making firms. This significantly improved the platform's liquidity conditions.
Whether it's in terms of capital reserves or compliance thresholds, these are real moats that platforms like Polymarket and leading platforms like Kalshi have built. Just a few months ago, Polymarket received a $2 billion investment from ICE, the parent company of the New York Stock Exchange, at an $8 billion valuation, and there are reports that it is now planning another round of financing at a valuation of tens of billions of dollars. Meanwhile, Kalshi has also completed a $300 million funding round at a $5 billion valuation. The two leading platforms now have substantial financial resources at their disposal.
The prediction market is currently a hot spot for innovation in the broader market, with numerous new projects emerging constantly. However, I'm actually not very optimistic about them. The reason is that the leading effect in prediction markets is actually stronger than many people imagine. How can new projects directly compete with leaders like Polymarket and Kalshi, who are continuously offering real money subsidies and are supported by high-level partners from the compliance world? How much capital do these new projects have to sustain such competition? While it's possible that some new projects backed by powerful sponsors can flood the market with funds, clearly not every project is in that position.
Haseeb Qureshi, the bald guy from Dragonfly, posted his predictions for 2026 a few days ago. He wrote a sentence saying:"Prediction markets are developing rapidly, but 90% of prediction market products will be completely ignored and gradually disappear by the end of the year."I don't know what his logic is, but I agree that it's not an exaggeration.
Many people are looking forward to a flourishing market with diverse opportunities and are eager to profit based on past experiences. However, such a scenario may be difficult to achieve. Instead of spreading bets thinly, it's better to directly focus on the market leaders.