Encirclement of Polymarket Bots

MarsBitNews

2025-11-24 13:27:18

null

Authors: Liu Kaiwen, EeeVee, BlockBeats

On November 22, 2025, a silent showdown is taking place on a prediction market for Polymarket.

On one side of the showdown is a mysterious trader named @totofdn. On the other side is an automated arbitrage robot named sunshines.

It all started with a trivial order. @totofdn placed a very small sell order: 5 units at $0.34. This action instantly compressed the market's bid-ask spread to less than $0.04—this is the magic number that triggers the platform's “order reward”.

Almost at the same time, the Sunshine reacted. A huge sell order was smashed into the order book: 100 copies No @ 0.34. The bot is coming, and it strictly follows the instructions of the code to earn liquidity rewards from the platform.

But it doesn't know that this is exactly the signal that @totofdn is waiting for.

@totofdn decisively consumed all the sell orders of the robot, pocketing 100 shares at an average price of 0.34. Meanwhile, the robot was forced to take over 100 shares at an average price of 0.66, completely unaware that it had fallen into a trap.

This is just the beginning. In the next four hours, this set of “fake hanging, real eating” combo moves was performed repeatedly. Sunshines acted like an out-of-control ATM, spitting out cash again and again. In 4 hours, dozens of repeated operations, over $1500. The robot's account was precisely emptied, while @totofdn remained unscathed and quietly left the scene.

This is a meticulously planned “cognitive siege” against automated scripts. It reveals a truth about on-chain arbitrage: here, automation does not equal intelligence; AI can enhance efficiency but can also lead to greater losses.

The “optimal solution” and “fatal flaw” of platform incentives

To understand the intricacies of this battle, we must first return to the rules of Polymarket itself. As a decentralized prediction market, liquidity is its lifeline. To incentivize users to provide depth to the market, Polymarket has designed a mechanism called the “Order Book Rewards Program.”

The core idea of this mechanism is simple: those who provide liquidity to the market will be rewarded. Specifically, as long as users place their limit orders within the blue maximum spread line (the so-called “spread”) in the designated market and meet certain share requirements, they can proportionally share in the reward pool provided by the platform. Rewards are automatically distributed at midnight every day, straightforward and direct. This spread is usually the current midpoint price ±3ct to 4ct, with the specific width set in real-time by Polymarket.

Once any rule is quantified, it will inevitably give rise to specialized “score brushing” strategies. The order rewards of Polymarket quickly attracted a group of special “miners”. They do not care about the outcomes of the predicted events themselves, only about how to efficiently obtain the rewards. Thus, automated arbitrage bots like sunshines have emerged.

The code logic of these robots is as follows:

Scan the market: Continuously monitor all markets that meet the reward criteria.

Price Difference Judgment: Check whether the current market's bid-ask spread is less than a certain threshold (e.g., $0.04).

Triggering a pending order: As soon as a price difference in a market meets the liquidity reward requirements, immediately place an order within the price difference range that complies with the reward rules.

Receive rewards: Wait for the reward distribution at midnight.

From a coding perspective, there is nothing wrong with this logic; it perfectly utilizes the rules. The robots tirelessly “fill the price gaps” across various markets, contributing liquidity data to the platform, and in return, they receive rewards from the platform. They are the “optimal solutions” to the rules and are regarded as “model citizens” by Polymarket.

But the problem is that these robots only analyze price differences, shares, and rewards; they do not understand market sentiment, opponent analysis, and they have no risk control. They cannot distinguish whether the sudden appearance of a small order that compresses the price difference to trigger conditions is a real trading demand or a carefully laid trap.

When @totofdn placed those 5 sell orders at No @ 0.34, the code from sunshines told it: “Opportunity has arrived! The spread has been compressed to 1¢, hurry up and place an order to earn rewards!” It had no idea that this 0.01¢ spread was fake and artificially created. It only saw the “optimal solution” of the rules, but didn't notice the “fatal flaw” behind this solution.

In the end, this robot born for rewards has become the prey of a more advanced hunter due to its mindless pursuit of rewards.

From physical warfare to cognitive warfare

From MEV to Jito, and now to the “robot hunting” of Polymarket, this smoke-free war of on-chain arbitrage is undergoing a profound evolution.

If the early MEV (Maximum Extractable Value) wars were a “physical battle” revolving around gas fees and block space, today's on-chain games increasingly resemble a “cognitive battle” that tests strategy and psychology.

In the wild era of MEV, victory belongs to those “scientists” who have the fastest networks, the strongest hardware, and the highest packing priority. They are like a group of trucks rampaging on a highway, using absolute power and speed to front-run, sandwich, and liquidate, extracting value from ordinary users' transactions. It was a simple and brutal era, where the competition was about whose “muscles” were more developed.

Subsequently, MEV solutions represented by Jito emerged, attempting to establish a new order in this chaotic physical war. By auctioning block space, Jito redistributes the profits from MEV, allowing validators and stakers to also share in the spoils. This alleviates network congestion to some extent, but also makes the acquisition of MEV more “legitimate” and “industrialized.” The war has moved from the shadows into the light, transitioning from individual heroism to an arms race among professional institutions.

The incident that occurred at Polymarket reflects that on-chain gaming has entered a new phase. What determines the outcome is no longer millisecond-level delays or exorbitant gas fees, but rather the understanding of the rules, insights into market players, and the application of strategies.

@totofdn did not use any advanced hacking techniques, nor did he mobilize vast computing resources. His only weapon was a deep understanding of the Polymarket incentive mechanism and an accurate prediction of automated script behavior patterns like sunshines. He won a war of information asymmetry and, even more so, a war of cognitive dimensions.

The rules of the on-chain dark forest are changing. Simple automated scripts, if lacking the ability to dynamically adapt to the environment and a sense of competition against opponents, will find it increasingly difficult to survive. They are like species that have not fully evolved; although they are highly efficient in specific ecological niches (such as farming rewards), once the environment changes or they encounter more advanced predators, they are completely defenseless.

From the physical battles of MEV, to the order battles of Jito, and then to the cognitive battles of Polymarket, on-chain arbitrage is evolving from a game for “engineers” into a game for “strategists” and “psychologists.” In this increasingly complex dark forest, only those participants who can continuously evolve and enhance their cognitive dimensions will ultimately survive.