Risks of grid & perp trading

Grid trading and perpetual futures trading are risky. You can lose the capital you trade with. Automation does not change that: a bot follows its rules faster than a human, including when the rules are wrong for the market.

This article is the overview. Each linked article goes deeper on one mechanism.

The plain version

• Trading can lose money. Bots can lose money. There is no configuration that removes market risk.
• Profits shown in previews or projections describe behavior inside a range, not a guarantee.
• Never trade with funds you cannot afford to lose entirely.

Grid-specific risks

A grid bot buys as price falls through its range and sells as it rises. Three things go wrong predictably:

Price exits below the range. Every rung on the way down filled a buy, so the bot holds inventory bought above the current price. The grid stops trading below its bottom rung, and the unrealized loss resolves only if price returns. This is the main way grids lose money.

The market trends instead of ranging. Grids monetize oscillation. A strong one-direction move either leaves the grid holding inventory (down) or watching from the sidelines after selling early (up).

Drawdown before profit. Even a grid that ends up profitable spends much of its life holding inventory at a paper loss. If that drawdown forces you to stop the bot at the wrong time, the paper loss becomes real.

Sizing matters more than configuration: a grid sized so a full-range fill is comfortable survives bad weeks; a grid sized to maximize action does not.

Perp-specific risks

Perpetual futures add three mechanisms spot trading does not have. Each has its own article:

• Leverage multiplies exposure relative to your collateral, and multiplies losses identically. More leverage means a smaller adverse move wipes out your margin.
• Liquidation is the venue force-closing your position when margin can no longer cover potential losses. It happens at the worst price by construction, and it is the standard way leveraged accounts die.
• Funding rates are an ongoing cash flow between longs and shorts. A position held for weeks can bleed a meaningful amount in funding before price moves at all.

How collateral actually backs a position, and the cross vs isolated distinction, is covered in Margin. And if you’re assuming low leverage means no risk, read Is 1x leverage safe?. The answer is more conditional than most people expect.

Anello’s grid and DCA bots run on Hyperliquid spot markets, where there is no leverage, funding, or liquidation. Perp positions you open on the terminal carry all three in full. That’s why these articles exist.

Operational risks

Automation depends on infrastructure: Hyperliquid’s API and websocket feeds, wallet signing, and correct configuration. Orders can be rejected by venue rules, networks can stall, and a wrong range or size is executed exactly as configured. Start small, verify the preview before launching, and check on a new bot’s first days rather than assuming.

Expectations, set honestly

A realistic mental model for a grid bot: many small realized gains while price ranges, against an occasional larger drawdown when it doesn’t. The strategy is a trade-off you choose, not an edge someone hands you. Fees also take their share of every fill; see fees and risks for the exact numbers.