Hyperliquid portfolio mode: one net position per token, many alerts

The problem portfolio mode solves

Run TT-Autotune alerts on several timeframes for several tokens and you quickly get conflicting signals: the 15-minute cell says long while the 4-hour cell says short. Executing each alert independently means opening and closing overlapping positions, paying fees on every flip. Portfolio mode treats the alerts as a sensor net instead and keeps one net position per token, adjusting it as the timeframes agree or disagree.

Portfolio mode is built for Hyperliquid. It is armed per exchange key, fails closed, and at the time of writing has placed no live orders: it runs in shadow and on testnet while it earns its acceptance record.

The timeframe ladder

4h

Sets direction for the token

1h

Sets the core position

15m

Adds or trims around the core

5m

Fine-tunes the final layer

reversal rule

A full reversal needs the 4h and 1h to agree

Risk brakes

The latch is deliberate. After a 10% drawdown the portfolio stays flat until you press resume, so a bad day cannot turn into a bad week while you are away.

margin ceiling

85% of account margin, never more

4% drawdown

Stops adding to positions

6% drawdown

Reduces to the core position

10% drawdown

Flattens everything and latches until you resume

token count

At most four tokens at a time

correlation

Tokens correlated 0.75 or more with one already held are excluded

How tokens are chosen

The preview ranks trained TT-Autotune cells for Hyperliquid-listed tokens and picks up to four that are not highly correlated. Live forward results only start boosting a cell's rank once it has at least 40 live trades, so a lucky handful of fills cannot promote a token. The extension shows the selection as a "Hyperliquid portfolio batch" card with the risk settings spelled out: 85% margin ceiling, 4% stop-adds, 6% core, 10% flatten and latch.

Arming, pausing and resuming

1. 1Preview the portfolio for a Hyperliquid key and review the selected tokens.
2. 2Configure it: tokens, margin and the risk settings above.
3. 3Arm it. Live arming requires explicit confirmation, a live active Hyperliquid key that passed preflight, shadow and testnet acceptance, and no active latch.
4. 4Pause at any time; the status changes to paused and no new adjustments are made.
5. 5Resume to clear a latch after reviewing what happened.

When to use it, and when not

Use portfolio mode when you want Autotune exposure on a handful of liquid Hyperliquid tokens without micromanaging overlapping alerts. Skip it when you want each alert to trade independently, when you trade other venues, or when you are testing a single strategy. Plain Batch Create alerts are simpler and easier to reason about.

Portfolio mode vs independent alerts

positions

Portfolio: one net position per token. Independent: one position per alert leg.

fees

Portfolio: fewer flips. Independent: every alert can open and close its own position.

control

Portfolio: rules decide adds and trims. Independent: each alert acts on its own.

venues

Portfolio: Hyperliquid only. Independent: any executable venue.

Why the gates are strict

Portfolio mode concentrates decisions: one bug or bad rule affects every token it holds. That is why arming requires shadow and testnet acceptance first, and why a latch after a 10% drawdown needs a human to resume. The gates follow the same fail-closed principle as the rest of TensorTrader's order path: when a confirmation is unexpected or a state is ambiguous, stop rather than guess.

Frequently asked questions

Does portfolio mode trade live today?

Not yet. It fails closed and runs in shadow and testnet until it passes acceptance; live arming has strict gates.

How many tokens can it hold?

Up to four, and never two that are highly correlated.

Keep reading

• Hyperliquid trading bot: how to automate Hyperliquid perps safely

  The complete guide to automating Hyperliquid: agent wallets, testnet, TradingView and GoCharting alerts, leverage and margin, fees and funding, portfolio mode, and the real results of our own testnet books.

• How TT-Autotune works: one alert, six regime champions

  TT-Autotune tunes the Lorentzian Classification strategy per token, per timeframe and per market regime, then syncs the winning settings into a single TradingView alert. Here is how the optimizer, the regime switch and the extension fit together, plus a live testnet ROI chart.

• DCA across timeframes: how margin splits by weight

  How TensorTrader’s DCA Across Timeframes spreads one strategy across a spectrum of chart timeframes, weights margin toward the base, and fits your alert quota.

• Hyperliquid leverage, cross margin and Unified Account explained

  How leverage and margin work on Hyperliquid perps, what Unified Account changes, and how TensorTrader sizes and protects automated positions.