NYY/BOSYankees ML+0.0¢LAD/SDDodgers ML+0.0¢PHI/ATLPhillies ML+0.0¢HOU/SEAAstros ML+0.0¢CHC/STLCubs ML+0.0¢TOR/WSHBlue Jays ML+0.0¢CLE/CINGuardians ML+0.0¢MIL/SFBrewers ML+0.0¢NYY/BOSYankees ML+0.0¢LAD/SDDodgers ML+0.0¢PHI/ATLPhillies ML+0.0¢HOU/SEAAstros ML+0.0¢CHC/STLCubs ML+0.0¢TOR/WSHBlue Jays ML+0.0¢CLE/CINGuardians ML+0.0¢MIL/SFBrewers ML+0.0¢

Proprietary tradingPrediction & event markets

POK Capital

We trade our own capital. Our systems price the same outcome on two venues and take both sides when the numbers don't agree.

Depthlive
spread

01 — Firm

We trade our own book.

We started in 2025. There is no outside capital, no clients and no fund behind us; the money at risk is ours.

Our market is exchange-traded event contracts, on venues like Kalshi and Polymarket US. Prices there are probabilities, which makes a disagreement between two venues something you can measure rather than argue about.

We are a small team and we intend to stay one. Everything that touches an order was written here, which is the only way we know to be certain what the system will do at four in the morning when nobody is watching it.

Founded
2025
Capital
Own balance sheet only
Venues
Kalshi · Polymarket US
Based
California & Ohio

02 — Strategy

The same outcome, priced twice.

Two venues list the same event and their prices drift apart. Different order flow, different fees, different people paying attention. When the gap is wider than what it costs us to trade it, we take both sides and hold the pair to settlement. We aren't predicting the game. We're trading the disagreement about it.

NYY/BOS · same outcomelive
Kalshi
Polymarket
Basis

shaded = below our cost to trade

01

Hedged at entry

We don't leg into a position and hope the other side is still there. Both sides go on together, or the trade doesn't happen.

02

Priced, not forecast

Nobody here has a view on who wins. The only question is whether the two prices add up to less than a dollar.

03

The bar is a number

Most of what the system sees, it turns down. Edge is thin and it goes away fast, so the decision to trade is arithmetic, not judgment.

03 — Platform

We built the whole stack.

Market data, pricing, execution, hedging and reconciliation are all ours. That is less a point of pride than a constraint: the edge lives in the gap between seeing a price and holding the position, and that gap isn't something you can outsource.

Order pathtoday
Data
Pricing
Execution
Reconciliation
Evaluated
Declined
Executed
DataBoth venues' books, normalised and timestamped
PricingEdge computed net of fees, slippage and hedge cost
ExecutionPaired orders with risk checks in the critical path
ReconciliationFills and settlements matched to venue truth

04 — Risk

Most of the code is there to stop us.

Every order passes limits that were written before the market opened. The interesting part of the system isn't the part that trades — it's the part that decides not to.

Limitsutilisation
Per-event exposurecap $250
Hedge completiontarget 100%
Unhedged residualcap 5 sh
Daily loss limitstop $40
01

Limits live in the code

Per-order, per-event and firm-wide caps, checked before anything reaches a venue rather than reported after it did.

02

It stops itself

A stale book, a fill we can't explain, a venue behaving oddly. The system halts and pages us instead of trading through it.

03

Nothing carries unhedged

Unmatched exposure is measured continuously and closed out. We don't hold a naked leg overnight to see how it turns out.

04

Reconciled against the venue

Every position is checked against venue records through settlement. If our books and theirs disagree, trading stops until we know why.