by Justin Lord · @Justin_lords
feb 2026

I studied how people pulled $40M from prediction markets last year.

Not betting. Not gambling. Math. Here's the full breakdown.

$40M+
Extracted in 12 months
86M
Bets analyzed
2.7s
Avg opportunity window

01 The basic mechanic

Prediction markets let you buy contracts on outcomes. "Will BTC go up in the next 15 minutes?"

A contract costs between 1¢ and 99¢. If you're right, it pays $1.00. If you're wrong, it pays $0. The price reflects probability — 70¢ means the market thinks there's a 70% chance.

Two platforms run these markets: Kalshi (regulated US exchange) and Polymarket (crypto-based). They price the same events differently because they have different users, different data feeds, different liquidity.

That's where the money is.

02 The arbitrage

When one platform prices "Up" cheap and the other prices "Down" cheap, you can buy both sides across both platforms for less than $1.00 total.

One side always wins. Winner pays $1.00. You pocket the difference.

THE CORE MATH

Kalshi UP: 38¢  +  Polymarket DOWN: 55¢  =  93¢

One side always pays $1.00 → guaranteed +7¢ per contract

03 Why "7 cents" is real money

$70
1,000 contracts
$350
× 5 trades/day
$10,500
× 30 days

04 The strike price edge

Here's what most people miss. Kalshi and Polymarket use different data feeds, so their "price to beat" (strike price) is slightly different. This creates a zone where both your bets can win simultaneously.

Kalshi
Price to beat: $66,000

Lower strike → buy UP

BUY UP38¢
Polymarket
Price to beat: $66,110

Higher strike → buy DOWN

BUY DOWN55¢

Where BTC can land:

WIN $1
⭐ BOTH WIN $2
WIN $1
Below both strikes Between strikes (bonus zone) Above both strikes

05 Every possible outcome

With the correct pairing (Kalshi UP 38¢ + Poly DOWN 55¢ = 93¢ total):

BTC landsKalshi UPPoly DOWNYou get
$66,200 (above both) ✓ $1.00 ✗ $0 $1.00 → +7¢
$66,050 (between) ✓ $1.00 ✓ $1.00 $2.00 → +$1.07
$65,900 (below both) ✗ $0 ✓ $1.00 $1.00 → +7¢

Zero scenarios where you lose.

06 The one rule you cannot break

THE PAIRING RULE

Get this wrong and there's a zone where both sides lose. Get it right and that zone becomes your bonus.

HIGHER STRIKE → BUY DOWN  ·  LOWER STRIKE → BUY UP

✓ Correct pairing

Kalshi ($66K)BUY UP
Poly ($66.1K)BUY DOWN
Between zoneBOTH WIN $2
Worst caseSTILL WIN $1

✗ Wrong pairing

Kalshi ($66K)BUY DOWN
Poly ($66.1K)BUY UP
Between zoneBOTH LOSE $0 💀
Worst caseLOSE EVERYTHING

07 Calculator — play with the numbers

38¢
55¢
$0 total 93¢ — profitable $1
$930
Cost
$1,000
Return
$70
Profit/trade
$10,500
Monthly

08 The honest part

Why most people fail at this:

Speed. Opportunities last 2-3 seconds. 73% of profits go to bots running under 100ms. Manual trading catches almost nothing.

One-leg risk. You need both sides to fill simultaneously. If one fills and the other doesn't — you're not doing arbitrage anymore, you're gambling.

Fees. Kalshi charges ~1-2% per trade. A 3% spread can become 0% after fees. Most retail-visible opportunities are 2-3%.

Capital. 3% return sounds thin until you're deploying $5K+ per trade. At $100 starting capital, profits are $2-$5 per trade.

Competition. ICE invested $2B in Kalshi. Hedge funds have dedicated prediction market desks. You're competing against million-dollar infra.

The edge: build a bot. One developer made $764 in a single day starting with $200 on BTC 15-min markets. The math works — you just need execution speed.