> For the complete documentation index, see [llms.txt](https://torch-1.gitbook.io/litepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://torch-1.gitbook.io/litepaper/continuous-market.md).

# Continuous market

Torch introduces a novel market structure designed for continuous forecasting of crypto token prices. Unlike traditional prediction markets (like Omen, Polymarket) that focus on discrete binary or categorical events, Torch allows traders to predict any price interval, at any time, for any future moment. The result is a unified dynamic market that builds a **time–price probability surface** – a collective signal layer of future price expectations.

Most prediction markets today are built around conditional tokens. These require a separate market to be created for each question and resolve based on an external oracle after a fixed date. While effective for event-based forecasting, these systems are limited in scope when it comes to price discovery, alpha detection, and continuous signal generation.

Torch replaces that model with range-based betting and deterministic resolution:

* Traders (human or AI) can stake on price ranges at any time, no need to create a new market
* Bets are evaluated at the specified future timestamp against the token price, fetched from price oracles
* All activity contributes to a shared forecast surface, revealing directional consensus, volatility bands, and speculative imbalance in real time

|          Feature         |                       Torch                      |            Prediction markets           |
| :----------------------: | :----------------------------------------------: | :-------------------------------------: |
|   **Market structure**   |       Continuous (price and time intervals)      |      Discrete binary or categorical     |
|   **Market structure**   |           Unified, perpetual per token           |         One market per question         |
|    **Betting format**    |      Range-based (e.g. $3500–3700 in 3 days)     |  Fixed outcome tokens (Yes/No, A/B/C…)  |
|    **Liquidity model**   |          Fully parimutuel, trader-funded         |    AMM-backed, requires external LPs    |
| **Resolution mechanism** | Oracle-based, objective token price at timestamp |   Oracle-based, often slow or disputed  |
|       **Use case**       |     Token price forecasting, alpha discovery     |    Event resolution, binary outcomes    |
|    **Forecast output**   |       Public probability map (price × time)      |     Outcome probabilities per market    |
|  **Participation mode**  |              Open and permissionless             | Requires market creation and LP funding |