A question no data feed can answer
Seer keeps a list of verified markets, and anyone can challenge a market on it. The challenger argued this one broke Seer’s policy: “Jesus” and “return” mean different things to different people, so nobody could say for sure how it resolves. The submitter argued the policy already covers that: the market resolves on credible reporting, and jurors don’t judge theology.
This is the kind of question where a jury matters. The jurors didn’t decide whether Jesus returns. They decided whether the market follows the written policy. Most markets never reach a court, and most disputes end in one round. This one used every step, so it shows how the whole process works.
Challenged the listing and opened the case. Argued the market is ambiguous (Article 4) and its answer is already known (Article 9).
Submitted the market and appealed after each loss. Argued it resolves on credible reporting (Article 18) about the most obvious meaning (Article 12).
From challenge to ruling
Submitted. The market is added to Seer’s verified list with a deposit.
Challenged. The challenger pays the jury fee. Case #717 opens and 3 jurors are drawn at random from people who staked PNK in the court.
Evidence. Both sides, and anyone else, post arguments for the jurors. 69 entries by the end.
Round 0: 2–1, don’t list. The losing side has a window to appeal.
Round 1: 6–1, don’t list. Appealed. The jury grows to 7 seats.
Round 2: 10–5, don’t list. Appealed. The jury grows to 15 seats.
Round 3: 21–4, don’t list. 31 seats, secret ballot: jurors commit a hidden vote, then reveal it.
Final. Nobody funds another appeal. The ruling is executed, the market stays off the list, and jurors and appeal funders are paid.
Each appeal doubles the jury, plus one
Drawn to scale by seats. A bigger jury makes the ruling harder to sway, and each round costs more, so appealing only makes sense for a side that expects to win.
What each side argued
Don’t list it
- Article 4, clarity. The policy rejects markets “subject to different interpretations”. Christian, Islamic, Jewish, Hindu and historical readings of “Jesus” and a “return” differ.
- No way to verify. If someone claims to be Jesus, who decides? There is no test a court can apply.
- Article 9, answer already known. Leaving belief aside, the historical figure died around AD 30–33.
- The right document. The stricter Verified Markets policy applies, not only the general resolution policy.
List it
- Article 18, credible reporting. If credible sources report a return it resolves Yes, otherwise No. Jurors don’t judge theology.
- Article 12, most obvious entity. Fringe readings don’t create ambiguity, or markets on “recession” or “war” would fail too.
- Article 9 bars past events, not future events someone thinks impossible.
- Precedent. Similar markets were accepted before.
All 69 entries, as posted
In the order they were posted. Open one to read it in full.
Paid to vote the way the ruling goes
Each seat puts PNK at stake. Jurors who vote with the final ruling keep their stake, earn the fee, and share the stake lost by the others. Jurors who vote against it, or don’t vote, lose it. The chance of being drawn grows with the PNK staked, so a juror can hold several seats.
Every round costs more
To appeal, both sides fund the next round, and anyone can contribute. The side that lost the last round pays 3× the new jury fee, the side that won pays 2×. The jurors’ fee comes out of that, and the rest goes to whoever funded the side that wins in the end.
All amounts in xDAI, the stable token used for fees on Gnosis Chain. Across the case, jurors were paid 552 xDAI in fees, and 16,704 PNK moved from jurors who voted against the ruling to jurors who voted with it.