Skip to content

Inputs and settings

The Backtester input form is a five-step funnel down the left side of the tool, plus an Advanced panel and a set of header controls. This page documents every field, its accepted values, and its default, then explains the validation rules that decide when a run can start. The Backtester is available to Free, Pro Lite, and Pro accounts.

Before you begin

  • Sign in to your account. The Backtester opens to Free, Pro Lite, and Pro members.
  • To pull holdings from a saved portfolio, build that portfolio first. For the steps, see Build your first portfolio.

Quickstart

  1. Open the form

    Go to the Backtester and work down the five steps from source to benchmark.

  2. Open Advanced only if needed

    Leave the Advanced panel closed for standard behavior, or open it to set commission, slippage, the rebalance method, drift tolerance, or a dividend tax rate.

  3. Clear any blockers

    Read the run preview. It lists every reason the run is still blocked so you can fix them together before pressing Run backtest.

What it shows

The form has five numbered steps, a collapsible Advanced panel, and header controls. The sections below describe each in order.

Inputs

Step 1: choose source

A dropdown sets where the tickers come from.

  • Custom (default): search and add individual tickers by hand. A search box accepts a symbol or a company name, and you can also paste a comma-separated list. Added tickers appear as a removable list with a "clear all" option. Equities and exchange-traded funds can be mixed freely.
  • Portfolio: pick one of your existing portfolios. Its holdings load automatically, and weights are set from the current market value of each position. A picker lists each portfolio with its position count.
  • Model portfolio: pick a pre-built model portfolio. Holdings and weights load the same way.
  • Combined portfolios: select several of your portfolios at once. Their holdings are merged into one basket through a multi-select picker with checkboxes and a running "X of Y selected" count.

If a chosen portfolio, model, or combination has no stock positions, a warning tells you to add positions first.

Step 2: set period and capital

  • Run name: a free-text label for this run. The default is "My Backtest". It is used in history and on the results header.
  • Start date: the simulation start. The default is five years before today. It cannot be earlier than the earliest supported date or later than the end date.
  • End date: the simulation end. The default is today. It cannot be later than today or earlier than the start date.
  • Initial capital: the starting dollar amount, entered with a dollar prefix. The default is 100,000. It must be greater than zero.

Step 3: choose weights

A dropdown sets how capital is split across holdings.

  • Equal weight (default): every holding gets the same share.
  • Custom percent: opens an Edit weights button that launches a modal where you type each holding's percentage. A running total shows in green when it sums to 100 percent and red when it does not. The modal restricts entries to two decimal places, offers an Auto-Normalize to 100 percent button when the total is close, and enables Save only once the total equals 100 percent.

Step 4: choose strategy

  • Strategy dropdown:
    • Static (default): hold the chosen weights, restoring them at each rebalance.
    • Dynamic: at each rebalance, rank the holdings by a chosen metric and hold only the top names. Selecting Dynamic reveals two more controls:
      • Metric: one of Lowest PE ratio, Lowest PB ratio, Lowest PEGY, Lowest Cash-Adjusted PEGY, Highest dividend yield, or Highest EPS.
      • Top N: how many names to hold each period. The default is 10, capped at the number of tickers.
  • Rebalance frequency dropdown: Never, Monthly, Quarterly (default), Semiannual, or Annual. This sets how often the portfolio is brought back to its target weights.
  • Reinvest dividends: an on/off switch, default on. When on, dividends are reinvested back into the portfolio. When off, they accumulate as cash.

Step 5: pick a benchmark

  • Benchmark: a ticker entered as text, default SPY, forced to uppercase. Any ticker can be the comparison line, for example SPY, QQQ, or VTI. For how the comparison reads, see Benchmark comparison.

Advanced panel

The Advanced panel is collapsible. Its defaults match standard behavior, so an untouched panel does not change the result.

  • Commission per trade: dollars charged on each trade. The default is 0.
  • Slippage: a decimal fraction of price lost to execution, where 0.001 equals 0.1 percent. The default is 0.001.
  • Rebalance method: Calendar (rebalance on the fixed schedule) or Drift (rebalance only when weights wander out of band). The default is Calendar.
  • Drift tolerance: how far weights may drift before a Drift-method rebalance fires, as a decimal where 0.05 equals 5 percent. The default is 0.05. It is enabled only when the method is Drift.
  • Dividend tax rate: a decimal applied to dividends only when reinvestment is off, where 0.15 equals 15 percent. The default is 0.15.

Header controls

  • New run: clears the current results and returns to the input form. It appears after a run exists.
  • Compare: switches the right panel to the multi-run overlay view. See Saved runs and comparison.
  • History: switches the right panel to the saved-runs list. See Saved runs and comparison.
  • A "remaining / total runs" counter appears only when a run limit is configured for your tier.

Outputs and how to read them

The form itself produces no figure. It produces a valid configuration that the run uses. The run preview on the right reflects your current settings and reports whether the run is ready or still blocked. When a run finishes, its report appears in the same panel. For the report, see Results and metrics.

How it is calculated

The form applies one piece of standard math directly: custom weights must sum to 100 percent before a run can start.

The weight total is the sum of every holding's percentage.

total percent = w1 + w2 + ... + wn

In this formula, each w is one holding's percentage and n is the number of holdings. The total must equal 100 percent. For example, three holdings at 40, 35, and 25 percent sum to 100, which is valid. If the total reads 98 or 103, the modal shows it in red and the Auto-Normalize to 100 percent button rescales each weight proportionally so the total reaches exactly 100. Entries are limited to two decimal places, so tiny rounding gaps are normalized rather than left to drift.

Data and timing

When you pull holdings from a portfolio, a model portfolio, or a combination, the current market value of each position sets the starting weights. Those values use AlphaVantage prices, with quotes delivered through AlphaVantage from Nasdaq, refreshed daily after the close. The Dynamic-strategy ranking metrics (PE, PB, PEGY, cash-adjusted PEGY, dividend yield, EPS) draw on AlphaVantage fundamental and price data.

Use cases

  • Pull a saved portfolio into a test in one click, rather than retyping every ticker.
  • Set custom weights to model a target allocation, then auto-normalize to clean up rounding.
  • Use the Advanced panel to add realistic trading costs and switch to drift-based rebalancing.

Limitations and disclosures

The form enforces input validity, not realism. The cost assumptions you set in the Advanced panel are estimates: the simulation applies your commission and slippage figures and does not model real order fills or market impact beyond them. A backtest is hypothetical and not a prediction.

Validation

The Run backtest button stays disabled until the inputs are valid. Blocking conditions include:

  • No tickers selected.
  • More than 100 tickers.
  • Initial capital of zero or less.
  • A missing start or end date.
  • A start date that is not before the end date.
  • Custom weights that do not sum to 100 percent.
  • In Dynamic mode, a Top N below 1 or above the ticker count.

All current blockers are listed together in the run preview, so you can fix them at once.

GNG Research provides equity research and educational tools, not investment advice. Nothing on the platform is a recommendation to buy or sell any security. Do your own research and consider your circumstances before making any investment decision.

What's next

What's next

Was this page helpful? For corrections or questions, get help and report an issue.

GNG Research provides equity research and educational tools, not investment advice. Nothing on the platform is a recommendation to buy or sell any security. Do your own research and consider your circumstances before making any investment decision.

© 2026 GNG Research. All rights reserved.