How much capital do you need to start?
Short answer: ~$500 on each exchange ($1,000 total) is the practical floor where cross-exchange arbitrage is worth it. Less than that and the $29/mo subscription takes too long to earn back. Long answer with the math below.
1 · Why you need capital on both exchanges
Cross-exchange arbitrage means buying YES on one side (e.g. Kalshi) and NO on the other (e.g. Polymarket) for the same event. Both legs are independent positions on independent venues — there’s no shared margin and no way to net them. You need enough capital sitting on each side to size into the leg there before the price moves.
Moving money between exchanges takes 1–3 business days for ACH-based platforms. By the time you’ve wired from one to the other, the arb you saw is long gone. So the working capital has to be pre-positioned on both sides, not a single pooled balance.
2 · What different bankroll tiers actually get you
The minimum capturable arb on Tellmarket is sized at $50. Most arbs fire with $500–$5,000 of top-of-book liquidity. You'll skip 80%+ of what you see because you can't put enough in to make the subscription back.
Practical floor. Split $250–$500 on each exchange. You can fully size into the smaller arbs (the $50–$500 range) and partial-size the bigger ones. At a typical 1–3% net edge, an arb-per-week cadence pays back the $29 in 4–8 weeks.
Comfortable. You can fully size most arbs that pass the gates. The $29/mo is a rounding error against the captured edge from 2–3 arbs a month.
At this size you start running into the top-of-book limit (gate of $50 is the floor, but most arbs cap around $5,000 fillable). You'll size into the full top-of-book on each leg and let the rest of your bankroll sit idle.
3 · The break-even math on the subscription
Tellmarket Pro is $29/mo. To make the subscription pay for itself you need to capture at least $29 of net edge per month. At a typical 1.5% net edge, that means you need to put roughly $1,933 of arb stake through the system per month:
That doesn’t mean you need $1,933 in capital — capital recycles as each arb closes. If you have $500 on each side ($1,000 total) and turn it over 2× a month on captured arbs, you’re at the break-even line. Higher edge or higher turnover both shrink the working-capital floor.
4 · What you don’t need
- A margin account. Both Kalshi and Polymarket are cash-only for event contracts. You’re posting the full notional on each leg, not a margin percentage.
- Crypto. Polymarket US runs on USDC but accepts USD on-ramp via debit card / ACH; you don’t need to hold crypto separately. Kalshi is straight USD.
- Trading software. Manual execution is the workflow — you get the alert, open the exchange, place each leg. No bots, no API trading required at this tier.
5 · The honest summary
~$500 per sideis the floor where this product makes economic sense. Under that, the subscription costs more than the arbs you can fully size into will return. Around $1,000–$5,000 per side is the sweet spot. Above $10,000 you’re going to leave capital idle waiting for arbs big enough to fully consume it. Don’t fund a Kalshi or Polymarket account with money you can’t leave parked for weeks — both exchanges have multi-day ACH withdrawal cycles, and yanking it on day 3 because you need rent defeats the whole point.