BestFolio vs AllocateSmartly: An Honest Comparison

If you are researching tactical asset allocation platforms, you have probably come across AllocateSmartly. They have been the established platform in this space since 2017, and their library is still the broadest in the industry. BestFolio is newer (we launched in 2025) and we built it because we wanted a few specific things that the existing tools did not give us. This post is an honest comparison of both. I am the BestFolio founder, so treat my opinions accordingly. The factual claims are checked against the live AllocateSmartly site as of the date below.

Last reviewed: 2026-04-27. Significant rewrite after factcheck feedback from a long-time AllocateSmartly subscriber on r/AllocateSmartly. Earlier versions of this post had several errors which are corrected below.

Quick comparison

If you only have a minute, here is the side-by-side. Detail follows below.

Strategy coverage

AllocateSmartly tracks more strategies than we do. Their list runs to roughly 95 published strategies, including the full Keller family (Bold, Resilient, Hybrid, Defensive, Vigilant, Classical, Protective, Elastic), the Faber catalog, ReSolve work from Adam Butler, the CSS Analytics suite from David Varadi, and several strategies that exist only on AllocateSmartly because the authors gave them privileged access to the rules. FMO3 and Grzegorz Link's GGCEM are two examples we cannot replicate from public sources, and we do not try to.

Our 47 published strategies cover the academic core: GEM, ADM, CDM, VAA, DAA, BAA, HAA, PAA, the Keller family we can build from public papers, Permanent Portfolio, Golden Butterfly, Risk Parity, Classic 60/40, and so on. Plus 13 fixed-allocation portfolios and a small number of original strategies we publish under the BestFolio Originals label.

Where we do something different is variant tracking. Each strategy in our database carries one canonical variant and often multiple parameter variants alongside it. A common one is "GEM with a 6-month lookback" next to "GEM with the canonical 12-month lookback." This is potentially data mining, and a long-time AllocateSmartly subscriber pointed this out fairly. Our defense is that we publish walk-forward and out-of-sample numbers for every variant, not just the parent. If a variant looks better than the parent only because we tuned it on the same window we are evaluating, the OOS column says so. We would rather show a variant that loses to its parent on OOS than hide it.

How a BestFolio signal is computed

Worth describing the pipeline because it is what we built the platform around. AllocateSmartly treats the signal as a black box for the user; we try to make every step inspectable.

• Daily ingest. ETF prices and dividends from Tiingo and yfinance, signals from FRED for macro inputs (Composite Leading Indicator, unemployment, yield curves).
• Universe filter. Each strategy carries its own asset universe from the paper that defined it. We do not improvise.
• Score and rank. Whatever the strategy says: 13612W momentum, breadth triggers, canary checks, score thresholds.
• Walk-forward gate. The strategy's historical performance is computed in walk-forward windows, typically 5-year in-sample to 1-year out-of-sample, rolled forward. The OOS curve and metrics are visible on every strategy page.
• Signal computation and email. End-of-month close, next-open execution. Subscribers see the next-month preview during the last few trading days.
• Blending. Combine multiple strategies into a portfolio with custom weights. Per-strategy contribution analysis shows where each strategy added or subtracted return and where the diversification benefit actually came from.

AllocateSmartly does most of these too. Their walk-forward methodology is stricter than ours in one important respect: they do not allow data tweaking inside the walk-forward window. We let users adjust the in-sample length, which is convenient for research but easier to abuse. If you care about that distinction, AllocateSmartly's approach is the safer default.

Signal delivery

Both platforms publish monthly signals at month-end close. Both let you preview the next month's expected allocation before the new month starts: AllocateSmartly via the alternate trading day setting on a custom portfolio plus the 09:30 ET expected-allocation view on the trading day, BestFolio via the next-month preview banner on each strategy page during the last several trading days. The mechanics differ, the capability does not. An earlier version of this post claimed signal previews as a BestFolio differentiator. That was wrong and is removed.

Both platforms send email notifications. Both let you customize when those emails fire by configuring the trading day on a custom portfolio.

Analytics and validation

Standard performance reporting is comparable on both: CAGR, max drawdown, Sharpe, Sortino, rolling returns, equity curves, correlation matrices, monthly heatmaps. If you only need these, neither platform is missing anything.

AllocateSmartly has the edge on tranching. They provide signal-day analysis that lets you compare end-of-month execution against alternative trading days for the same strategy, and they expose a tranching workflow where you split a strategy into multiple sleeves trading on different days. We do not have that, and most users do not use it, but if you do it is a meaningful gap on our side.

BestFolio adds three things on the analytics side that we use ourselves and that came out of questions we wanted to answer that the existing tools did not handle:

• Walk-forward visible per strategy. The OOS equity curve is the default view on every strategy page, not a hidden tab. Every metric we publish exists in two flavors: full-history canonical and OOS-only. If they disagree, we want the reader to see that immediately.
• Variant comparison. Side-by-side parameter alternatives with their own OOS metrics, so you can see whether "GEM with a 6-month lookback" actually held up out-of-sample or only looked good in-sample.
• Blend contribution analysis. For a multi-strategy portfolio, attribute each percentage point of return to the strategy that produced it, and decompose the diversification benefit by correlation contribution.

Pricing

AllocateSmartly: $49 per month, $399 per year, or $499 per year for Pro. Free tier with three sample strategies. Lifetime price-lock guarantee for paid subscribers as the platform grows.

BestFolio: $29 per month, $299 per year. Free tier with six strategies plus full backtest data and equity curves, no credit card required. We do not have a Pro tier; everything paid is the same plan.

We are cheaper. AllocateSmartly is more polished and has eight more years of refinement in the same product. If you find an existing subscription is delivering for you, the price gap is probably not enough by itself to switch. The free tier comparison is the place to spend five minutes if you want to test fit before paying anything.

UCITS and EU positioning

An earlier version of this post implied that AllocateSmartly does not cover UCITS-eligible ETFs. That was wrong. AllocateSmartly has had UCITS coverage for a long time.

What we do specifically: every tactical strategy on BestFolio carries a UCITS-eligible ETF mapping for the EU subscriber, and we backtest that mapping rather than only the US tickers the original paper used. This matters if you are an EU investor and the gap between the paper's SPY and your VWCE-equivalent UCITS proxy adds up over 30 years. We were built EU-first in that sense; it is a positive claim about our priorities, not a negative claim about AllocateSmartly's coverage.

Historical data depth

AllocateSmartly's real-data history is longer for several assets than ours. We use 30 years of real ETF data where it exists and synthetic extensions for older periods on assets where the ETF did not exist yet (TLT before 2002 is mocked from CRSP-style bond returns, GLD before 2004 from spot gold, and so on). On our charts the synthetic portion is marked. If you want the longest possible real-data history with no synthetic extension, AllocateSmartly is the better source.

What AllocateSmartly does better

This list comes mostly from a long-time subscriber's email; it is more credible than what was here before.

• Track record. Live signals since 2017. Eight market environments of public history including 2018, 2020, 2022, 2025. We have one full year.
• Originated strategies. FMO3, Grzegorz Link's GGCEM, and others where the author granted AllocateSmartly access to the unpublished rules. These are not on BestFolio and never will be.
• Walk-forward methodology. Stricter than ours: no data tweaking allowed inside the walk-forward window. Better default for users who do not want a footgun.
• Tranching and signal-day analysis. Tooling for splitting a strategy into multiple sleeves trading different days, with comparison views. Not many users need it but if you do it is real.
• Polish. Eight years of UI iteration. Some users will find their interface more intuitive than ours; some will find ours more direct. Try both before deciding.
• Established community. Long-running discussion forum and a subscriber base that has worked through enough market cycles to recognize edge cases.
• Real-data history. Longer for several assets, no synthetic extension on the older periods.
• Underperformed Watchlist concept. Introduced by AllocateSmartly. We shipped our own version after they did and we should have led with that attribution. We have updated our feature page accordingly.

What BestFolio does better

• Variant tracking with OOS. Multiple parameter variants per strategy, each with walk-forward and OOS metrics so you can tell tuning from real signal.
• Visible walk-forward per strategy. OOS curve as the default view, not a hidden tab.
• Blend contribution analysis. Attribute return and diversification benefit to specific strategies inside a blend.
• UCITS-eligible mappings on every tactical strategy. Pre-mapped EU ETF proxies that are backtested with the actual EU asset, not just the US ticker.
• Free tier depth. Six strategies with full backtest data and equity curves vs. AllocateSmartly's three sample strategies.
• Lower paid price. $29/$299 vs. $49/$399.
• Free tools open to non-subscribers. A factor-lens regression tool, a UCITS finder, and a drawdown analyzer that anyone can use without an account.

Which should you choose?

If you want the broadest published TAA library, the longest real-data history, AllocateSmartly's exclusive strategies (FMO3, GGCEM, others), or the strictest walk-forward methodology, AllocateSmartly is the right tool. If you have an existing AllocateSmartly subscription you are getting your money's worth from, the price gap is not enough by itself to make a switch worthwhile.

If you want variant-level walk-forward visibility, blend contribution analysis, UCITS-eligible mappings on every tactical strategy, a deeper free tier to test before paying, or simply a cheaper paid plan, BestFolio covers those cases.

The signals on overlapping strategies should be nearly identical. Both platforms apply the same paper rules to the same end-of-month closes. Differences come from execution timing, ticker choices on older history, and dividend handling, and they show up at roughly 0.2 to 0.5 percent per year. The decision between platforms is not the signal; it is the workflow and the analysis tooling.

Many serious TAA investors run both. The signals cross-validate, and the analytical lenses are different enough that having both perspectives is a real advantage. If you are evaluating for the first time, start with the AllocateSmartly free tier and the BestFolio free tier, spend an hour on each, and pay for whichever workflow you actually want to live in.

Changelog

2026-04-27: Significant rewrite after factcheck feedback from a long-time AllocateSmartly subscriber. Corrections: removed false signal-preview differentiator, removed false email- customization claim, dropped US-first framing in favor of EU-first claim about us, dropped binary UCITS claim in favor of specific surface, added attribution for Underperformed Watchlist concept, replaced vague "modern UX" claims with concrete features, added comparison table, added pipeline section, expanded AllocateSmartly strengths section, added honest historical-data depth note. Strategy count updated to ~95 (AS) and 47 (BestFolio).

2026-04-10: Earlier draft (now superseded).