How Index Providers Create Synthetic Pre-Launch Histories | Index One

In this edition of Index One Insights by Index One, we examine how index providers create pre-launch performance histories using a process known as backtesting, or backfilling.

📖 How Index Providers Create Synthetic Pre-Launch Histories

🎥 Optimizing Portfolios Through Well-Structured Index Blends

📊 What’s your view on the current stock market rally?

➡️ Subscribe to our bi-weekly newsletter to stay updated with current news in the index industry, insightful articles and new Index One features.

To receive our newsletters via email, sign up here.

How Index Providers Create Synthetic Pre-Launch Histories | Index One

When a new investment index launches, it is standard practice to present a historical performance record alongside it, often spanning years or even decades. Much of that history predates the index’s official launch date and is constructed through a process known as backtesting or backfilling.

Understanding how this works, and where its inherent limitations lie, is an important context for anyone evaluating an index fund or ETF.

What Is a Synthetic Index History?

A synthetic (or reconstructed) index history is a hypothetical performance record created by applying an index’s current rules to historical data. In other words, the provider takes today’s methodology and asks: if this index had existed in 1995, how would it have performed?

Index providers are transparent about this practice. Their disclosures typically state that all information for an index prior to its launch date is hypothetical back-tested, not actual performance, based on the index methodology in effect on the launch date.

This kind of reconstruction is common across the industry. A 2023 Morningstar study of 782 indices found that the median index had 10.5 years of backtested performance data- in other words, far more pre-launch history than live history.

How Is It Done?

Providers use several methods to fill in the gaps before an index officially existed:

• Applying rules retroactively: The index’s selection and weighting criteria are run against historical market data. If an index today selects the top 100 companies by dividend yield, the provider calculates what that basket would have looked like each year going back in time.
• Prepending proxy data: When a new ETF tracks an existing index that predates it, providers often attach the underlying index’s returns to the ETF’s own history. For example, the i1 US 500 (Market Cap) index only launched in 2024, but the index itself has data stretching back much further. Analysts can prepend the older index data, adjusted for fees, to create a longer synthetic record.
• Statistical simulation: For newer, more complex indices, providers may use models to reconstruct plausible historical returns, using techniques like Monte Carlo simulation or factor decomposition.

Known Limitations of Backtested Data

Reconstructed histories are a useful tool, but they come with well-documented limitations that investors should understand.

• Selection bias and data snooping: When designing an index, there is a risk that methodology parameters are tuned to those that happened to perform well historically. Researchers David Bailey and Marcos López de Prado found that pre-launch backtest studies of ETFs released between 1994 and 2014 showed average index outperformance of around 5% per year. Once those funds launched, actual excess returns were approximately 0%.
• Pre-launch performance inflation: The Morningstar study found that new indices outperformed their category benchmarks by an average of 1.4% per year in the five years before fund launch, almost entirely on the back of backtested data. That figure dropped to just 0.39% in the first year after funds went live.
• Survivorship bias: Historical reconstructions may inadvertently include only companies and assets that exist today, omitting those that failed, went bankrupt, or were removed from markets. This can inflate annual returns and cause risk metrics like the Sharpe ratio to appear more favourable than they truly were. Rigorous index providers seek to account for this by incorporating full historical constituent universes into their reconstructions.

What Regulators Say

Regulatory frameworks in major markets set clear expectations for how hypothetical performance is disclosed:

• The SEC’s Amended Marketing Rule (effective November 2022) requires clear disclosures that hypothetical data is not actual performance, was created with hindsight, and cannot predict future results.
• FINRA prohibits broker-dealers from using backtested performance in public communications, citing concerns about misleading investors.
• Industry bodies such as the Standards Board for Alternative Investments (SBAI) recommend that investors always ask providers to separate simulated performance from live, realised performance, and to disclose the rules and assumptions underpinning the reconstruction.

What Investors Should Consider

Backtested histories offer useful context, particularly for newer strategies, but they should be evaluated alongside live performance data. Some practical questions worth asking:

• How much of the track record is live versus backtested? A fund with ten years of history but only six months of live performance is very different from one with ten years of real returns.
• When was the index launched relative to the fund? Morningstar found that one in six funds began tracking an index that had been live for fewer than 30 days, meaning investors were essentially relying entirely on backtested data.
• What assumptions were made? Good providers disclose whether transaction costs, fees, liquidity constraints, and other real-world frictions were incorporated into the reconstruction. Many are not.
• Does the methodology appear well-timed in hindsight? If an index claims to have navigated every major market crisis perfectly in backtests, that warrants further scrutiny.

The Bottom Line

Synthetic pre-launch histories are a standard and, when used rigorously, a valuable part of how modern indices are built and communicated. They help investors understand how a strategy would have behaved across different market environments, information that is genuinely useful when evaluating a new product.

At the same time, a backtested track record is not the same as a live one. Transparent providers make this distinction clear, disclose the assumptions behind their reconstructions, and present backtested data alongside, not instead of, live performance wherever it exists.

As always, past performance, including backtested and hypothetical performance, is not indicative of future results.

This information is for educational purposes and does not constitute investment advice. Index One Limited does not sell, promote, or sponsor any financial instruments.

Optimizing Portfolios Through Well-Structured Index Blends

By April 2025, third-party model portfolios in the U.S. reached about $7.96 trillion, reflecting growing adviser reliance on pre-built, diversified frameworks using index funds and ETFs. U.S. ETF assets surpassed $11 trillion by late 2025, with advisor allocations expected to reach 25.5 % of client assets by 2026, overtaking mutual funds in popularity.

Index blends are now widely used across the investment industry: wealth managers and advisers use them in model portfolios, institutions rely on them for strategic benchmarks, robo-advisors implement them in automated ETF portfolios, and retirement funds use them for diversified benchmarks. Multi-asset and passive investing have made index blends a standard tool for portfolio construction and performance evaluation.

A notable example from BX Partners is the U.S. Large Cap Model, a blend of two phaseinvest indices calculated on Index One’s platform. By combining a Size-based Large Cap Core Index with a Dividend Growth & Yield + Momentum Index, the model offers U.S. large cap exposure across different market segments.

In this webinar, we will explore:

• The adoption of index blends in investment portfolios.
• The creation and methodology of the BX is the U.S. Large Cap Model index blend.
• How the indices were built and automated on the Index One platform.
• The application of the index blend in investment portfolios.

Register now

POLL: What’s your view on the current stock market rally?

In June, inflation remains elevated as earlier energy cost increases continue to work their way through the economy, while lower oil prices have helped improve sentiment.

What’s your view on the current stock market rally?

💡 Vote now!

Vote now

Your trusted partner for end-to-end index calculation solutions.

Index One offers a trusted, holistic index calculation platform that delivers fast, flexible, and scalable indexing solutions. We help build indices based on custom strategies and provide capabilities for backtesting, rebalancing, corporate actions, and distribution.

Create a custom index or enhance your current indexing workflows with our comprehensive backtesting, maintenance, dissemination and reporting indexing solutions.

Learn more.