Skip to content

Fed decision cross-venue arb

Paper bot

Buys both sides of one Fed outcome across Polymarket and Kalshi when the pair costs under $1 after fees.

Forward test2 trades recorded · 0 settled

Simulated paper trading only — orders are not sent to Polymarket or Kalshi.

Fed decision cross-venue arb returned +0.08% from its first trade on 13 Nov. Buying the S&P 500 returned −0.71%, measured from its 12 Nov base close. +0.79 pts vs the S&P 500.

All recorded dates · 13 Nov – 16 Nov

1W, 1M and 3M unavailable: only 2 days recorded so far.

What $100 in Fed decision cross-venue arb became, against the S&P 500The chart starts with the first trade on 13 Nov. The S&P 500 comparison uses its 12 Nov base close, the last trading close before that trade. 0 trade marks.What $100 became$99.4$99.6$99.8$100.014 Nov15 Nov16 NovStrategy+0.08%S&P 500−0.71%
Inspect a date with the arrow keys, or drag across the chart.
Drawdown for Fed decision cross-venue arbHow far Fed decision cross-venue arb fell from its own running peak between 13 Nov 2026 and 16 Nov 2026; worst −0.11% on 13 Nov 2026.Below its own peak0%−0.11%Worst−0.11%
The strategy line is green when it is ahead of the S&P 500 and red when it is behind; amber is the same $100 in the S&P 500 over the same dates. Selected ranges rebase both lines within the selected dates. The drawdown band uses recorded chart observations and keeps earlier running peaks. The facts below cover the lifetime record. No trade marks are recorded in this snapshot.
Return
+0.1%Lifetime · from $1,000.00
vs S&P 500
+0.8 ptsLifetime · S&P base close 12 NovS&P −0.7% from that close
Equity
$1,000.84Paper account value
Trades
20 settled
Max drawdown
0.1%Lifetime · worst fall from a peak
Days live
12Since the paper record began

In plain words

TL;DR

Forward test: buys YES on one venue and NO on the other for the same Fed decision, only when the pair costs at least 1¢ a share under $1 after fees, and holds both to the result. On 2026-11-04 the December 'no change' pair cost $1.02, so it may rarely or never trade.

How it bets

Each hour the bot reads Kalshi's Fed-decision series, takes the next two meetings closing more than 24 hours ahead, and matches each outcome to Polymarket's 'Fed Decision in <month>' event by exact title. For each outcome it prices both ways to own it: Polymarket YES + Kalshi NO, or Polymarket NO + Kalshi YES. Either pair pays $1 a share however the Fed moves, so buying it below $1 after both venues' taker fees locks in the difference. It fetches the Polymarket order book only when the best asks already leave at least 1¢ a share after fees, buys $50 of the pair in whole shares (both legs the same count), skips a gap over 15¢ as a stale or mismatched quote, and holds both legs to resolution. Forward test, unproven; it may rarely or never trade.

Exact rules

Paper bot on Kalshi’s Fed-decision series and Polymarket’s ‘Fed Decision in <month>’ event. Every hour it takes Kalshi’s next two meetings closing more than 24 hours ahead and finds each Polymarket event by slug prefix and end date (never a fixed slug, which carries a timestamp), then pairs outcomes by exact title: 50+ bps decrease, 25 bps decrease, no change, 25 bps increase, 50+ bps increase. For each outcome it prices both ways to own it (Polymarket YES + Kalshi NO, or Polymarket NO + Kalshi YES); either pair pays $1 a share however the Fed moves, so a pair under $1 after both taker fees locks in the difference. It fetches the Polymarket order book only when the best asks plus both fees can still leave at least 1¢ a share. A pair trades when it costs at least 1¢ a share under $1 after fees, both legs’ asks sit inside 2–98¢, and the gap is at most 15¢ (a bigger gap is a mismatch or a stale quote); $50 buys the pair in whole shares, both legs carry the same share count, at most 3 pairs a UTC day, one per outcome, held to resolution. Polymarket’s fee comes from the market’s own fee schedule; Kalshi’s is 7% × shares × P × (1 − P) rounded up to the cent. A Kalshi leg pays $1 a share when the result is its side and nothing for the other side; a void refunds the entry cost. Risks: the venues word their resolution rules separately, so the same outcome can settle differently on each, and a paper fill buys both legs at one instant while live the seconds between them move the price. Forward test, unproven; there is no single source for textbook two-venue arbitrage.

Bets

Record 0–0Realized $0.00Unrealized +$0.89Open 2

Price is per share; a share pays $1 only if both legs win. The fee is Kalshi's taker fee, charged on each leg as 7% of shares × price × (1 − price) rounded up to the cent, and the two legs add up.

Price is per share; a share pays $1 if it wins. The fee is Polymarket's taker fee for the whole bet. Polymarket links include our referral code.

Signed amounts show gains and losses. Open amounts are unrealized; settled results retain their recorded status.

2 settled and open bets on this record.

  • +$0.71

    Unrealized

    Open · Now worth $4.53

    7¢ a share · now 9¢ · $3.65 + $0.17 fee · pays $50.00

  • No

    +$0.18

    Unrealized

    Open · Now worth $45.50

    90¢ a share · now 91¢ · $45.00 + $0.32 fee · pays $50.00

Show raw fills · 2 recorded

All 2 recorded fills available on this page are shown (up to 100).

  • Will the Fed increase interest rates by 50+ bps after the December 2026 meeting?

    $45.00

    notional

    no at 0.900 · 50 shares · $0.32 fee

  • Will the Fed increase interest rates by 50+ bps after the December 2026 meeting?

    $3.65

    notional

    Yes at 0.073 · 50 shares · $0.17 fee

Last hourly run

Considered 15 · Trades 0 · Skipped 16

Skips are normal when the market fails the bot's published filter, the price is outside its band, the book is empty, or the full stake cannot fill.