The chart is not your account.

One percentage can hide several different stories

A strategy chart rises. Your account does less well, or loses money over what looks like the same period. The natural question is whether the software failed. Sometimes an execution problem needs attention. Sometimes the comparison itself is wrong. Without knowing what each number measures, those possibilities are difficult to separate.

A backtest asks what a defined model would have done under historical assumptions. A forward model follows rules as new data arrives but can still use simulated fills. A live account contains actual orders, costs, deposits, withdrawals and decisions made by its owner. Each can be useful evidence. They are not the same record.

This distinction matters even when a platform uses real market data. Data can be live while the trading record is hypothetical. A label should explain the calculation and execution basis, rather than leaving the reader to infer it from a moving chart.

The execution gap is a real engineering problem

Alpaca’s paper-trading documentation identifies several things its simulation does not reproduce, including market impact, latency-related slippage and queue position. It also describes other differences between the simulated environment and live trading. A successful paper run therefore tests only part of the journey. Read the broker’s explanation of paper-trading limits.

Consider an illustrative order for a security quoted at $50. A model may assume a fill at its chosen reference price. A live order reaches the broker later, joins the available liquidity and might be filled in pieces. If your buying power is insufficient, the broker may reject it altogether. The resulting account cannot be expected to trace the model exactly.

This example is not evidence that a particular provider delivered poor execution. It shows why an investigation needs timestamps, order status and fill information. “The chart is different” identifies a question; it does not yet identify the cause.

Sizing can change the portfolio you actually hold

Suppose a strategy targets a 5% position. In a hypothetical $20,000 allocation, that is $1,000 before other constraints. In a $1,000 allocation, it is $50. If the instrument costs $200 and the broker does not permit fractional shares for that order, the smaller account cannot express the target exactly. That is a sizing constraint, not a new view about the market.

Existing holdings can change the required trade as well. If you already own the security, a system calculating the difference between current and target positions may need fewer new shares. A manual trade made after activation can change the next calculation again.

SquaredQ’s Reader accounts for allocation, NAV, holdings, prices and broker capabilities when turning shared targets into account-specific quantities. Shared strategy instructions do not imply identical positions or fills across subscribers. Review the signal-to-order explanation before using one account as a benchmark for another.

Match the measurement before judging the result

Start with the dates. An account activated halfway through a month does not have the same opportunity set as a model measured from the first day. Next check the strategy version, product mix, allocation method, leverage assumptions and whether a risk stop interrupted execution.

Then separate trading results from cash movements. If an account begins with $10,000, receives a $5,000 deposit and ends with $15,300, the 53% increase in balance is not a 53% investment return. A meaningful return calculation needs to account for the size and timing of cash flows.

SquaredQ’s account dashboard uses NAV and NAV-derived metrics. Those whole-account figures should not automatically be interpreted as cash-flow-adjusted, strategy-only returns. The performance disclosures explain this boundary. Broker records remain necessary for reconciling actual account activity.

A practical way to investigate a difference

Keep the research document and version you evaluated. Record the activation time, broker environment and allocation settings. For the period in question, compare expected activity with the orders and executions in your broker account. Look for rejected orders, permissions issues, session interruptions, manual trades and risk actions.

A single missing order may be operationally significant without explaining the entire difference. Conversely, a clean execution log does not prove that a strategy’s economic thesis is sound. Separate the operational question-did the configured process run?-from the investment question-did that process produce an acceptable result?

When you contact support, provide the time window and an appropriate order reference through the designated support channel. Avoid sending API secrets, passwords or full unredacted account screenshots. Specific evidence helps the team trace a problem without making assumptions about your whole portfolio.

How SquaredQ helps you make the comparison

The Data Room provides the research starting point. Portfolio Studio provides a separate space for hypothetical combinations. MyAlgo provides the execution configuration. Your broker provides the account record. Keeping those roles distinct is SquaredQ’s answer to the confusion that starts when every chart looks like a personal result.

That structure does not eliminate slippage, model error or trading losses. It gives you a more useful way to ask what happened. Begin with the Data Room guide, then follow the evidence through to the live account instead of assuming the smoothest curve is the right comparison.