Poly Research & Robotics

A dedicated community for people actively building trading bots, trading strategies, and application layers on top of prediction markets.

[Poly Research & Robotics @PolyR_Robotics]

Introducing Poly Research & Robotics.

A free Discord community for people building trading bots and applications for prediction markets. Whether you’re just getting started or already running a profitable strategy, you'll find helpful tools, resources, and information in our server.

https://t.co/UB4SW1gDBg

[Poly Research & Robotics @PolyR_Robotics]

We're building trading bots for prediction markets in public.

Join our free Discord that has:

> Free historical datasets✅
> Free logging & wallet monitoring tools✅
> Trader analysis reports✅
> Active community helping each other✅

Bring ideas. Ask questions. Learn. Find your edge.

👉https://t.co/2dusiIW527

[Poly Research & Robotics @PolyR_Robotics]

@k0k1eth @0xd1namit @0xHonz Please answer @0xd1namit’s question regarding polymarket global, who’s the house?

[Poly Research & Robotics @PolyR_Robotics]

@0xd1namit We felt the same way and started an entire community dedicated to polymarket bot building, and application development, 100% free!

https://t.co/MMtoljaudM

[Grazka @grazkag]

Don't listen to this nonsense from people who are just engagement farming.

There will be no @Polymarket IPO on December 31 2026.

TradingView automatically puts in that date if nothing is officially announced, unless you actually believe all the companies in the screenshot are having an IPO on that exact same day?

[Poly Research & Robotics @PolyR_Robotics]

@poly_data About 7

[mahera @mahera777]

.@Polymarket dropped a new feature

All important news is displayed directly on the market chart

You can immediately see what events influenced the change in odds

[Poly Research & Robotics @PolyR_Robotics]

@CarOnPolymarket Same.

[Poly Research & Robotics @PolyR_Robotics]

@devjoshstevens @PolymarketDevs I have a secret:

(they weren’t going anywhere)

[Josh @devjoshstevens]

If ghostfill issues drove you away from Polymarket, whether up/down markets or any markets I completely understand. But please give it another shot. It’s now working better than ever, and if there’s anything in the UI that could make your experience even better, we want to hear it. We’re building the best defi prediction product user-first, one improvement at a time.

[Poly Research & Robotics @PolyR_Robotics]

Reverse Engineering the Polymarket HFT Bot That Made 112K in 23 Days

Trader: Slip-Me

Quick rundown:

He is a bot running on Polymarket, trading only the 5M Bitcoin Up or Down market.

>Around 8,500 trades per active day (no sells)
>Over 188K buys across 23 days

From the orderbook side, it looks like a textbook passive market maker quoting both Yes and No on every market he touches. THAT IS THE DISGUISE.

THE STRATEGY:

The actual strategy is reactive directional trading.

1. He opens both legs of every 5M market within the first few seconds.
2. Then he watches the BTC tape for 3–4 minutes and loads more shares onto whichever side is winning.
3. By the close, his position is usually 2x to 5x heavier on the eventual winner.
4. Then he holds every share to expiry and the resolution oracle pays him out.

No model or predictive indicators, just reacting to what BTC is doing in real time.

> Net for the month: +$111K on $4.3M in deployed capital
> ~2.6% ROI
> 22 of 23 days green, only one losing day in the entire dataset

What I found interesting:

The spread leg of his book actually loses money.

His average paired cost is around $1.02, so every time he gets both legs filled he is already underwater by 2 cents per pair.

That spread leg bleeds about 97k for the month. The directional leg pays him +$204K which more than covers it.

The real edge is in the late-window loads on the dominant side

If you were going to attempt to copytrade him, you would want to filter to just these markets (5x or heavier skew = 99.8 percent dominant-side win rate, 1 loss out of 520 markets at that level).

Infrastructure Stack (best guess):

> Binance WebSocket for the BTC feed, deepest liquidity, lowest latency, the de facto reference for crypto market makers.
> Chainlink BTC/USD oracle for resolution truth, Polymarket's 5M BTC markets settle on Chainlink at the close.
> Polygon RPC through Alchemy or QuickNode, public RPCs would rate-limit him to death at this throughput. Order signing happens inside the bot itself (or pre-signed in batches, like a lot of bots do), so the RPC is mostly just for state reads and reconciliation.
> VPS, likely AWS (EC2) or Hetzner in Amsterdam or Ireland, sits close to the Polymarket CLOB (which lives in London).

The hard part is the algorithm…

1. Keep both legs balanced as the BTC price and orderbook move.
2. Estimate fair value BEFORE entering. Slip-Me's $1.02 paired cost is barely above true fair, most homemade bots land at $1.05 or worse, which is the line between making money and losing it.
3. Make the right call in the last 90 seconds on when to go directional on the dominant side vs. when to back off because BTC just reversed.
4. Handle the throughput. At 8,500+ trades a day across 140 markets, you're processing a TON of data events firing in parallel, and the bot can't silently drift or break at 3 a.m.
5. Stay fast on both legs. When balancing two sides of a market, even a few seconds of lag can leave you lopsided, you'll lose the arb and get stuck buying a lot more on one side to drag your VWAP back into line.
6. Do NOT add take-profit logic. The whole thing depends on hold-to-expiry settlement, and the economics break the moment you start selling shares mid-window.

Want more of these reports on Polymarket traders?

Join our Discord! 👉 https://t.co/UB4SW1gDBg

[Poly Research & Robotics @PolyR_Robotics]

@husky_vs Husky the local weatherman

[Poly Research & Robotics @PolyR_Robotics]

Full 10-phase report, methodology, and replication notes:

https://t.co/B8VGdqERXO

Join our free Discord for free reports, datasets, and tools for prediction market bot building.

https://t.co/vdeFRmTjyf

[Poly Research & Robotics @PolyR_Robotics]

This Polymarket wallet printed $119,586 in 28 days, not by finding bad odds, but by entering before the orderbook caught up.

> +$119,586 profit
> +82.95% ROI on $144K deployed
> 26 of 27 active days closed green

Full breakdown below👇

[Poly Research & Robotics @PolyR_Robotics]

His SELL volume exceeded his BUY volume by $72,735 before settlement was even counted.

The pattern:
→ Buy one side when it's cheap vs spot → Wait 15-60 seconds for the order book to catch up → Sell aggressively into the move

This SELL engine is what turned fast entries into locked-in profit.

[0xEmoni @0xEmoni]

One week after Polymarket’s migration to $pUSD:

+$320M in $pUSD
410K holders
12.3M daily transactions

These numbers are outstanding for just one week...

@Polymarket is hitting a new ATH almost every day

[Poly Research & Robotics @PolyR_Robotics]

@Damir_Akaza Gravia is great.

[Saurav @saurav_tweets]

polymarket is running a full scale prediction market with 2,500+ live markets right now

hyperliquid is still testing with just one

the real question -> is hyperliquid even a prediction market yet?