# Prediki

A wiki with a market attached — anyone writes the question, edits it, votes the result.

*https://predictionmarkets.tools/tools/prediki · Forecasting Platforms*

## Facts

### At a glance

| Field | Value |
| --- | --- |
| Vendor | Prediki |
| Category | Forecasting Platforms |
| Job | forecasting |
| Website | https://www.prediki.com |
| Pricing model | subscription |
| Free tier | true |
| Open source | false |
| Licence | none |
| Self-hosted | false |
| Tested hands-on | false |
| Last updated | 2026-09-19 |

### Pricing

| Tier | USD | Period |
| --- | --- | --- |
| Free | 0 USD | — |
| Essentials | 10 USD | — |
| Professional | 5 USD | — |

### Availability

| Field | Value |
| --- | --- |
| Jurisdictions | global |
| Open to US persons | true |
| KYC required | false |

### Markets

| Field | Value |
| --- | --- |
| Subjects | politics, macro, business, science, sports, crypto |
| Settlement | play_money |
| Resolved by | The question owner enters the final result against the judgement rule written on the question; if they do not, any user may propose the result, and a proposal that finds a supporter goes to a 72-hour vote weighted by invested capital. |

### Economics

| Field | Value |
| --- | --- |
| Taker fee | none |
| Maker fee | none |
| Liquidity model | amm |
| Platforms | web |
| AI features | none |

### Interfaces

| Field | Value |
| --- | --- |
| API | false |
| WebSocket | false |
| Scripting | none |
| Python | false |
| MCP server | false |
| Export | none |

### Capabilities

Yes: charting, live_trading, portfolio_tracking, calibration_scoring, alerts

No: screening, order_book, backtesting, automation, paper_trading, cross_venue, news, tax_reporting

*Verified: pricing 2026-09-19; availability 2026-09-19; markets 2026-09-19; capabilities 2026-09-19; economics 2026-09-19.*

## What it is

A prediction market built like a wiki. Anybody may post a question, and anybody else may then
rewrite its title, propose a better judgement rule, add answer options, write the background
article, or propose the final result — each change going through a support-and-vote procedure
with a 72-hour window rather than through a moderator. The market is the part that scores you;
the editing is the part that makes it unusual.

Trading is in **Credits**, a free virtual currency, with nothing to deposit and nothing to
withdraw. You invest Credits for or against an answer, and the price moves immediately as a
result — there is no order book and no counterparty to find, so a thin question still has a price.
Prices across the answers of a question are normalised to sum to 100%, which is what makes the
multi-band numeric questions readable: an S&P 500 question on the site carries six ranges pricing
out to 1.9%, 2.0%, 11.2%, 71.6%, 9.6% and 3.7%.

The record you build is **expertise points**, awarded per question and consolidated per topic, so
the platform's claim is not "here is your Brier score" but "here is the topic you are actually
good at". Questions with a larger incentive award more points. There is no published scoring
formula, no calibration chart and no Brier or log score anywhere on the site.

Money reaches forecasters through **Incentives**: a sponsor pays a sum in advance, the incentive
runs for a defined period, and at settlement it is apportioned among the traders — the better and
earlier the prediction relative to other traders, the larger the share. Prediki's own help pages
describe the mechanism in exactly those words and nowhere quantify it.

## Availability

Open worldwide, free to join, no identity check and nothing to deposit. The terms require you to
be at least 18 and legally capable of entering a contract, and Prediki reserves the right to
refuse, suspend or terminate access at its sole discretion. Nothing in the terms singles out a
country.

The operator is Prediki Prediction Markets GmbH, an Austrian limited liability company registered
at the Commercial Court of Vienna on 30 December 2011, with contact addresses in Vienna, Kiel and
San Francisco. The interface runs in English, German, Russian and Chinese, and the terms name
both prediki.com and prediki.de as the site.

## Pricing

Two products on one platform, and only one of them is what this card is about. Public prediction
markets are free with unlimited users. Confidential questions — visible only to users the owner
invites, under a non-disclosure agreement — are the paid tier: Essentials at USD 10 a month for
up to ten users and at most ten concurrent confidential questions, and Professional priced per
participant per month at USD 5 for 11 to 100 users, USD 4 for 101 to 200, USD 3 for 201 to 500
and USD 2.50 above 500, capped at a hundred concurrent confidential questions. Academic
institutions get 50% off, non-profits 90%, and annual prepayment buys two months.

There is no trading fee and no cost to a forecaster at all. Recruiting a crowd, consulting and
training are quoted separately.

## Markets & resolution

Topics are whatever users have written, and the topic tree is correspondingly long-tailed —
politics, macroeconomics and elections next to boxing, motorsport, astronomy and the
Russo-Ukrainian war. The public questions open on 19 September 2026 were a university course
market: a set of S&P 500, WTI crude and college-football questions tagged *WFU PM Fall 2026* and
owned by a named academic, with 41 to 52 transactions each and a few hundred views.

Resolution starts with the question owner, who is prompted automatically when a question falls
due and enters the final result against the judgement rule on the page — the S&P 500 question
names a specific quote page and the exact reference moment, 18 September 2026 at 16:00 EDT. The
interesting half is what happens when the owner does not turn up. Prediki marks the question
*Publication is overdue*, and any user may then propose the result themselves; a proposal needs
one supporter to open a 72-hour vote, votes are weighted pro rata to capital invested in the
question, and a vote against must be explained in a written comment. That is a genuine answer to
the author-resolution problem, and this catalogue has not found another platform that implements
one.

The regulations also require that a result be empirically observable and measurable, that answer
options be exclusive and exhaustive, and that a modification never alter a question's substance.

## Integrations

There are none. No API, no webhook, no export button, no first-party client, and nothing in the
help pages that addresses automation. What you can take instead is the licence: user-contributed
content on public questions is published under CC BY-SA 3.0, which is a more permissive position
on reuse than any other platform in this category takes — Good Judgment Open forbids automated
extraction outright and Hypermind's panel rules treat live prices as confidential. Prediki gives
you permission and no mechanism.

## Limitations

**Liveness is the thing to check before you invest a season here.** The site works, the price
charts update and questions settle, but the footer copyright ends at 2022, the charting library
is Highstock 1.3.2, and the only public questions open in September 2026 came from one university
course. Several help articles — Credits, Price, Real-money distribution — are empty stubs reading
"Please start an article". This is a maintained platform rather than a developed one.

The scoring is opaque. Expertise points are described but not defined, incentive shares are
described as depending on how early and how precise your prediction was without a formula, and
nothing on the site produces a number you could put in front of a research audience the way a
Brier or log score travels.

Editability cuts both ways. A question you traded can be rewritten by other users, subject to a
vote, and Prediki's own legal page says content is not subject to editorial supervision and that
virtually all of it can be modified by anyone at any time.

And the incentive economy depends entirely on sponsors showing up. No sponsor, no money, and the
questions open today carry none.

## Alternatives

[Manifold](https://predictionmarkets.tools/tools/manifold) is the same "anyone writes the question" idea with far more traffic,
a play currency, and the author rather than a vote deciding the result. [Metaculus](https://predictionmarkets.tools/tools/metaculus)
if you want staff-written questions and a scoring rule you can cite.
[Hypermind](https://predictionmarkets.tools/tools/hypermind) if the contest prize is the point, though its questions are written
for sponsors rather than by you.

## FAQ

### Is Prediki real money?

Trading is not. You trade in Credits, Prediki's free virtual currency, and there is nothing to deposit. Real money enters only as an Incentive — a sponsor pays a sum in advance, and it is apportioned to traders at settlement according to how early and how precisely they predicted.

### Who resolves a question on Prediki?

Its owner, against the judgement rule written on the question page. When an owner lets a settlement go overdue, any user may propose the final result; a proposal needs one supporter to open a 72-hour vote, and votes are weighted by capital invested in the question.

### Who runs Prediki?

Prediki Prediction Markets GmbH, a limited liability company under Austrian law registered in Vienna since December 2011, with listed offices in Vienna, Kiel and San Francisco. The interface is available in English, German, Russian and Chinese.

### Can I reuse Prediki question data?

Content on public questions is user-contributed and published under CC BY-SA 3.0, which is unusually permissive for this category. Confidential questions are excluded, and there is no API or bulk export to take it out with.

## Background

- [How accurate a prediction market's price actually is](https://predictionmarkets.tools/guides/how-accurate-the-price-is.md) — What research establishes about reading 73 cents as a 73% chance — where market prices are calibrated, the deviations that recur, and what the horizon does.
- [Scientific and government forecast hubs](https://predictionmarkets.tools/guides/scientific-forecast-hubs.md) — Public-health agencies run open, scored forecast hubs anyone can join. The unit of submission is a model, not a person, and nothing there is a personal record.
- [What a forecasting platform's score actually measures](https://predictionmarkets.tools/guides/what-a-forecasting-score-measures.md) — Proper rules, the two Brier conventions that differ by a factor of two, crowd-relative scores that do not travel between sites, and why profit is not accuracy.

## Also worth comparing

- [Manifold](https://predictionmarkets.tools/tools/manifold.md) — Anyone can open a question, anyone can take a side, and the currency buys nothing.
- [Metaculus](https://predictionmarkets.tools/tools/metaculus.md) — Proper scoring and public track records on questions nobody can take a position in.
- [Social Science Prediction Platform](https://predictionmarkets.tools/tools/social-science-prediction-platform.md) — Forecast what a study will find, before it finds it, and be scored against the estimate.
- [Confido](https://predictionmarkets.tools/tools/confido.md) — Open-source forecasting workspace you host yourself, with a Brier score pointing upwards.
- [Fatebook](https://predictionmarkets.tools/tools/fatebook.md) — Write down what you think will happen, in Slack or a browser, and get scored on it.
- [Good Judgment Open](https://predictionmarkets.tools/tools/good-judgment-open.md) — Brier scores against the crowd, and the recruiting ground for Superforecasters.