MarketPawns — White Paper

Version: 1.5
Based on the implemented MarketPawns architecture

---

SECTION 1. Problem and Vision

Markets are not random, but the vast majority of participants lack the tools to systematically extract profit.

Classical trading is characterized by three fundamental vulnerabilities that prevent traders from achieving consistent results:

1. Subjectivity of decision-making. Market participants enter and manage positions based on psychological factors (fear, greed) rather than strict probabilities.
2. Static trading algorithms. Traditional trading advisors (Expert Advisors) operate according to rigidly deterministic rules, which excludes the possibility of adapting to a changing market context.
3. Lack of continuous re-evaluation. A standard algorithm makes a decision exclusively at the moment of entering a position. If the nature of the price movement changes, the algorithm passively waits for the execution of Stop Loss or Take Profit orders.

The MarketPawns Architectural Solution
The MarketPawns system performs continuous re-evaluation of every active position upon reaching key price levels. Every action on the platform is based on the Expected Value (EV) metric. If the market context transforms and the trade's EV becomes negative, the trading agent algorithmically initiates the closure of the position.

---

SECTION 2. Market Models — The Basic Unit of Analysis

Market noise is not a signal. The system exclusively processes structures that possess formalizable characteristics.

The analytical core of MarketPawns is based on the evaluation of structural market models, rather than classical indicators. A model within the system is a dynamic structure of price movement, characterized by its internal geometry and evolution over time.

Each model is formalized through:

• Distinct geometry: pivot points selected according to the project's proprietary algorithms, calculated key levels, and a statistically justified target.
• Measurable proportions: ratios of price movement amplitudes and time intervals.
• Calculated "target" level: a coordinate to which the price statistically gravitates during the development of this structure.
• Hundreds of numerical characteristics: formalizing its context, formation history, and current state.

Chess Analogy: In chess, the value of a pawn (its potential) is not a constant — it algorithmically increases with each successful advance along the files towards the promotion square. Similarly, a market model in MarketPawns does not have a static value at the moment of its discovery. The system re-evaluates the model's potential as it passes each new price milestone: the further the price moves according to the scenario, the higher the certainty and the more accurately the platform predicts the final outcome.

The MarketPawns database aggregates many millions of such models across key financial instruments and timeframes, forming a fundamental dataset for training machine learning algorithms.

---

SECTION 3. Entry Decision Technology

The neural network architecture generates exclusively probabilities. The final trading decision is made based on a strict mathematical formula.

3.1. Analysis via Model Ensemble

For each identified market model, the system calculates several hundred numerical features: from the proportions of local price movements to cluster affiliation and the context of adjacent models.

An ensemble of specialized AI models (Deep Learning and Gradient Boosting) performs parallel processing of these features. Each model in the ensemble solves a highly specialized task:

1. «Will a trend reversal occur from the resistance level?»
2. «Will the price reach the first target level?»
3. «Can the price reach the second, deeper level?»
4. «Are there signs that the model's potential is exhausted and the trade should be closed immediately?»

The result of each model's work is the probability of a specific event occurring (a number from 0 to 1).

Attestation "passports" of models: Each ML model is equipped with an attestation passport containing evidence-based metrics of its accuracy on historical data. This ensures transparency of probabilistic assessments and eliminates the "black box" effect.

3.2. Expected Value (EV) Filter

The raw probability is integrated into the Expected Value (EV) mathematical filter:

EV = P(win) × Profit − P(loss) × Risk

Where:

• P(win) — the probability of success, verified by the neural network based on the historical accuracy of the model (from the passport) in a similar confidence range.
• Profit — the distance from the entry price to the target level (in capital).
• Risk — the distance from the entry price to the scenario cancellation level (Stop Loss).

Imperative rule: A trading signal is retransmitted to the broker's terminal exclusively under the condition EV > 0.

3.3. Algorithmic Risk Management

In addition to probabilistic assessment, the system implements strict algorithmic capital control. When forming each order, the trade volume is calculated mathematically so that the potential loss upon reaching the scenario cancellation level (Stop Loss) strictly does not exceed the value allowed by the given strategy (e.g., a fixed percentage of the available balance). Risk management is embedded in the entry calculation formula itself and prevents deposit "overload".

---

SECTION 4. Trade Lifecycle — Multi-Level Control

Opening a position initiates a new phase of analytical control. The trading agent is activated each time the price reaches a significant level.

MarketPawns implements cyclical re-evaluation at key stages of holding a position:

Level 1: Entry

The order is placed at the mathematically optimal price. The system algorithmically chooses between aggressive positioning at the approach level and conservative positioning (at the calculated level), maximizing the EV metric.

Level 2: Intermediate Target Level (Bumper)

Upon reaching the intermediate target, a recalculation is initiated: «Is it mathematically reasonable to hold the position?» If fixing the profit provides a higher EV, the trade is closed.

Level 3: Pullback to the Entry Level

In case of a price pullback to the entry point, the trade is closed algorithmically via a trailing-stop system. This process is provided by built-in risk management modules for capital protection; neural networks do not participate at this stage.

Level 4: Reaching the First Target (Target 1)

Upon fulfilling the initial plan (Target 1), the system recalculates the EV to make a decision on whether to fully fix the profit or hold a portion of the position until the next targets.

---

SECTION 5. Infrastructure and Scalability

Technological stability is the foundation for implementing algorithmic strategies.

• Independent Trading Agents: The platform allows launching multiple autonomous agents. Each agent can operate on a completely separate, isolated strategy or effectively combine multiple strategies simultaneously without causing conflicts.
• Parallel Processing: AI computations are distributed across a cluster of Python workers. Dozens of market models are analyzed simultaneously, but the architecture is fully scalable — throughput is limited exclusively by the availability of computing power and the volume of data available for analysis.
• Multi-Broker Support: The platform is natively integrated with MT4 and MT5 terminals, providing isolated and parallel management of trading accounts.
• Monitoring: A web panel broadcasts the status of agents, balances, and active orders. The EV decision log provides the ability to audit the agent's logic at any historical moment.
• Fault Tolerance: Local buffering ensures the preservation of EV data during temporary losses of connection with terminals, preventing the loss of trading commands.

---

SECTION 6. Sandbox — Research Environment

Sandbox is an interactive environment for visual analysis of market models and hypothesis validation without risking real capital.

Analytical functionality includes:

• Visualization of the geometry of discovered models on the price chart.
• Detailing of entry levels, cancellation zones, and a cascade of targets.
• Retrospective analysis (Backtesting) of historical model performance.

---

SECTION 7. Tokenomics (PAWN-COIN) and Prediction Markets

The PAWN-COIN token is the base unit of account of the platform's internal economy.

The value of the ecosystem is formed based on:

1. The effectiveness of algorithmic trading.
2. The accuracy of collective analytics (community forecasts).

7.1. Internal Prediction Markets

The fundamental mechanism of interaction with the community is implemented through the already launched Backed Forecasts.

Participants form forecasts regarding the outcome of specific market scenarios (for example: «Will a Reverse occur?», «Will the price reach the P6 level?»).

• The forecast is backed by a Stake in PAWN-COIN tokens.
• The stake is algorithmically converted into a position (YES/NO) on the internal prediction market.
• Result verification is fully automated — the system records actual price movement and algorithmically closes the markets without moderator intervention.
• A correct forecast generates a financial Payout and increases the user's Influence Score.

7.2. Emission Mechanisms (Proof-of-Contribution)

Users generate PAWN-COIN tokens by contributing to the ecosystem:

• Successful Forecasts: Systematic profit extraction on Prediction Markets.
• Content Creation: Developing educational materials, publishing video analytics, and expanding the project's presence in the information space.

7.3. Utility

PAWN-COIN is designed to evolve from an internal market collateral instrument into the economic coordination layer of the MarketPawns ecosystem. In the current product architecture, the token already sits at the intersection of market participation, contribution rewards, reputation growth, wallet progression, and auditable financial accounting.

This creates a utility model tied to real platform activity rather than abstract token circulation:

• Market Utility: PAWN-COIN powers Stake-based participation in Backed Forecasts and serves as the accounting unit for payouts, refunds, and settlements.
• Contribution Utility: the token can be emitted for actions that directly improve the ecosystem, including educational content, analytical reviews, translations, case labeling, and QA for new scenarios.
• Reputation Utility: Influence Score tiers provide a trust layer for scaling rewards, prioritizing qualified contributors, and gating access to higher-value participation mechanics.
• Progression Utility: wallet growth reflects not only successful forecasts, but also the user's cumulative contribution and status inside the platform.
• Audit Utility: every emission, adjustment, payout, refund, and future governance-related balance movement can be recorded in an auditable ledger.

As the ecosystem scales, this utility layer can expand through:

• Bounty Marketplace: structured token payouts for measurable product-improving work.
• Governance-Lite: limited token-based voting on new market types, feature priorities, bounty categories, and Sandbox tools, while the trading core remains outside direct token governance.

In this model, PAWN-COIN is not a standalone speculative asset. It is the internal economic interface through which participation, contribution, reputation, and coordination are aligned as the platform grows toward a decentralized economy.

7.4. Token Supply and Allocation Framework

To support ecosystem growth, contributor incentives, treasury resilience, and disciplined market formation, the tokenomics model uses a fixed supply of 1,000,000,000 PAWN-COIN.

The high-level allocation framework is structured as follows:

Bucket	Share
Community	35%
Treasury	20%
Team	15%
Liquidity	10%
Reserve	15%
Advisors	5%

The Community allocation functions as the main growth and participation layer of the ecosystem and is broken down into:

Community Bucket	Share of Total Supply
Influence Rewards	15%
Prediction Rewards	10%
Airdrop	5%
Referral Program	5%

7.5. Treasury Structure and Initial Circulation

The Treasury allocation is designed as a dual-control strategic pool:

• Founder-Controlled Treasury: 10% of total supply
• DAO Treasury: 10% of total supply

This 50/50 structure is intended to preserve early execution capacity while allowing a gradual transition toward broader ecosystem coordination.

For a conservative staged launch, the recommended initial circulation is 12% of total supply. This level is designed to be large enough for early market formation, community activation, and trading functionality, while remaining disciplined enough to avoid unnecessary early supply pressure.

A logical initial composition may include:

• 5% from the Liquidity allocation for initial market formation
• 2% from the Airdrop allocation for the first community activation wave
• 2% from Prediction Rewards and Influence Rewards for early ecosystem participation
• 1% from the Referral Program allocation for controlled growth activation
• 1% as an initial strategic ecosystem distribution pool
• 1% for launch-aligned treasury-supported operational incentives

7.6. Unlock Principles and Vesting Logic

The unlock design follows several core principles:

• Community allocations are distributed gradually to support long-term ecosystem growth.
• Team allocations are subject to multi-year vesting schedules.
• Treasury reserves are intended for long-term development and strategic initiatives.
• Reward emissions prioritize meaningful participation, forecasting performance, and ecosystem contribution.
• Liquidity allocations are released progressively to support healthy market formation.
• No large-scale unlock events are planned that could create unnecessary market pressure.

For a conservative staged launch, the recommended vesting logic is:

• Team: 12-month cliff followed by 36-month linear vesting
• Advisors: 6-month cliff followed by 24-month linear vesting
• Founder-Controlled Treasury: initially non-circulating, with milestone-based releases tied to development, partnerships, and strategic execution
• DAO Treasury: initially non-circulating, activated progressively alongside governance-lite and community-approved ecosystem programs
• Reserve: initially locked and released only for strategic, defensive, or long-horizon ecosystem needs
• Prediction Rewards / Influence Rewards: emitted gradually through measurable platform participation over a multi-year horizon
• Airdrop: partially unlocked at launch for initial community activation, with the remaining portion distributed in controlled waves
• Referral Program: released progressively based on verified user growth and ecosystem-quality acquisition
• Liquidity: seeded at launch for market formation, with any remaining allocation added progressively based on real market depth and trading conditions

7.7. Launch Readiness Logic

MarketPawns approaches token readiness as an execution sequence rather than as a marketing event:

1. Internal utility framing of PAWN-COIN across market participation, contribution rewards, reputation, accounting, and ecosystem coordination.
2. Formal public tokenomics publication covering supply, allocation, treasury structure, unlock principles, and vesting logic.
3. Public token structure announcement defining the market-facing launch structure.
4. TGE after legal and liquidity readiness once the project has completed the necessary external launch preparation.

---

SECTION 8. Transparency and Verification

Trust in the system is based on the possibility of independent mathematical control.

• Retaining the full history of EV snapshots ensures the transparency of every trading decision.
• The resolution of forecasts on Prediction Markets is algorithmic, completely eliminating the possibility of manipulating outcomes.

---

SECTION 9. Roadmap

The development of the MarketPawns ecosystem is structured across the following key stages:

✅ Phase 1: Core Development (Implemented)

• Data Aggregation: Creation of a fundamental database storing millions of market models.
• AI Core: Development of an ensemble of specialized AI models (Deep Learning and Gradient Boosting).
• Decision Making: Integration of the mathematical EV filter for probability assessment.
• Risk Management: Implementation of algorithms for multi-level control of open positions.
• Infrastructure: Construction of a high-load, scalable architecture based on Python workers.
• Integration: Implementation of connectors to popular MT4 and MT5 trading platforms.
• Gamification: Launch of the internal Prediction Markets to engage the community.

🔄 Phase 2: Community Scaling (Current Stage)

• Infrastructure: Migration of neural network computations to the Kubernetes architecture.
• Author Attraction: Deployment of Bounty campaigns for content creators (YouTube, articles).
• Audience Growth: Aggressive growth of the user base stimulated by Prediction Markets mechanics.
• Liquidity: Increasing participant engagement on internal prediction markets.
• New Strategies: Development of strategies using additional types of technical analysis models.
• Token Utility Expansion: Extending PAWN-COIN beyond market collateral into contribution rewards, wallet progression, and reputation-linked participation mechanics.
• Bounty Marketplace: Launching structured token payouts for educational content, case labeling, translations, analytical reviews, and QA around new market scenarios.
• Token Readiness Framework: Formalizing the public tokenomics framework, treasury structure, unlock principles, and staged launch sequencing for the future public crypto layer.

🔜 Phase 3: Decentralization and Finance (Planned)

• Governance-Lite: Introducing limited token-based product voting for new market types, feature priorities, bounty categories, and Sandbox tooling without granting direct control over the trading core.
• Public Token Structure: Announcing the market-facing token structure after the tokenomics and launch-preparation framework are finalized.
• Tokenization: Transitioning the internal currency (PAWN-COIN) into a full-fledged Crypto Coin on a public blockchain through a staged sequence that culminates in TGE after legal and liquidity readiness.
• DeFi Fund: Creation of a decentralized investment fund for capital management based on the platform's confirmed AI models.
• Analytics: Creation of an advanced analytical platform based on aggregated data.