Why 90% of My Polymarket Bot Trades Happen in the Last 3 Minutes

 I did not plan this. I noticed it in the data.

When I started logging every trade my bot made I expected the activity to be spread out across the full market window. Buy a little here, sell a little there, collect the spread all day. That was the theory. The reality looked completely different. Almost everything was happening right at the end.

At first I thought something was wrong with the bot. Maybe it was slow to start. Maybe the early market conditions were too wide to trade profitably. I dug into the logs and what I found actually made sense once I thought about it.

Early in a market the probability is uncertain. On a BTC 5 minute market nobody really knows where price is going in the next 300 seconds. The spread is wide because nobody wants to commit. The order book is thin. My bot would post orders and nothing would fill because everyone else was doing the same thing - waiting.

As the market gets closer to resolution something changes. The probability starts to collapse. If BTC has been flat for 4 minutes and the market is asking whether it will be above a certain level in 60 seconds, the answer is becoming clearer. Traders start taking positions. Volume picks up. The order book fills in. Suddenly there are real counterparties on both sides.

That is when my bot gets its fills.

The last 3 minutes of a 5 minute market is where the liquidity actually lives. The spread tightens. The volume spikes. The probability is moving fast enough that there is real edge in being on the right side of the book at the right moment. My bot is built to sit and wait for exactly this window and then move quickly when it opens.

This taught me something that changed how I think about market making on prediction markets generally. You are not just competing on price. You are competing on timing. Posting orders at the wrong time means sitting in a dead book waiting for fills that never come. Posting at the right time means catching a wave of volume that moves through fast and fills both sides cleanly.

The bot does not trade any less in slow periods. It just does not get filled. The trades that actually execute are concentrated in that final window almost every time.

I now build every new strategy around this observation. When does liquidity actually appear in this market. That question matters more than what price to post at.

If you are building on Polymarket and your fill rate feels low, check when your fills are happening. The timing pattern in the data will tell you more than the price logic ever will.

All my code is open source at github.com/casatrick

Comments

Post a Comment

Popular posts from this blog

How a Polymarket Arbitrage Bot Actually Works: 5 Strategies Explained

A Polymarket arbitrage bot scans prediction markets for pricing inefficiencies and executes trades to capture the resulting spread. That's the concept in one sentence, but arbitrage on Polymarket isn't one strategy, it's several distinct mechanisms, each exploiting a different kind of mispricing. Here's a breakdown of how each one works, based on an open-source implementation. Intra-market arbitrage is the core mechanic. In a binary Polymarket market, YES and NO token prices should always sum to $1.00. When combined pricing falls below that after fees, buying both outcomes locks in a profit that doesn't depend on which outcome actually happens. Buy YES and NO, merge the tokens into a complete set, redeem for $1.00 per share. This is the closest thing to genuinely risk-free arbitrage available in prediction markets. Combinatorial arbitrage extends the same logic to multi-outcome markets, like price-range or bracket-style markets with three or more possible outcomes...
Read more

Why Every Polymarket Trading Bot Needs to Adapt to TWAP

When building a Polymarket trading bot, it's easy to focus on prediction models, latency, and order execution. However, platform infrastructure changes can have an even greater impact on a trading system than a new strategy. With Polymarket introducing TWAP (Time-Weighted Average Price) settlement for crypto Up/Down markets, I spent the last few days reviewing how my bot behaves near market resolution. Supporting TWAP wasn't simply replacing one price calculation-it required rethinking assumptions around settlement, execution timing, and trade evaluation. This update reminded me that trading bots aren't static software. They're living systems that must evolve alongside the markets they interact with. Over the coming weeks, I'll share a detailed technical breakdown covering the architecture changes, implementation decisions, and lessons learned while updating my Polymarket trading bot for the new TWAP model.
Read more

Building a Polymarket Trading Bot: Order Book Monitoring, Slippage Handling, and Position Sizing

  Most first-attempt Polymarket bots place a market order and call it done. That works fine on liquid, high-volume markets, but Polymarket has thousands of long-tail markets with thin order books and wide bid-ask spreads, where a naive market order can fill at a price far from the quote seen just a moment earlier. Polymarket's API is split into two services. The Gamma API handles public, read-only market discovery. The CLOB API handles order books, prices, and actual trading. Reading market data requires no authentication, while placing orders does. A solid bot workflow starts by discovering a market and pulling its order book, then calculating edge by comparing a model's estimated probability against the market's implied probability. Position sizing should use a fractional-Kelly approach, capped by a hard maximum exposure per market, since that cap matters more for risk control than the Kelly fraction itself. Before filling an order, checking remaining order book depth ...
Read more