Twenty US-listed stocks, equal-weighted at 5% each, plus a small basket of commodity ETFs. Nothing else — no options, no futures, no private assets, no crypto. Every holding is an ordinary listed security you can see and price at any moment.

By a fully rule-driven quantitative process, developed and validated over more than two decades of market history before ever running live. The same inputs always produce the same portfolio — no earnings forecasts, valuation opinions, macro calls, or discretionary overrides. The underlying signal logic itself is proprietary and not published.

Names are rebalanced roughly monthly, with typical turnover of 20–30% — most months replace four to six positions. Exposure, though, adjusts daily: the stock list can stay identical for a month while gross exposure moves substantially.

They decide how much to hold, separate from what to hold. The regime dial reads how broadly stocks are participating in the market trend — healthy participation means fully deployed, deteriorating participation pulls the book toward cash. The volatility target compares the portfolio's own realized volatility to a target level — calm book, exposure rises; turbulent book, exposure falls. Multiplied together they set the day's gross exposure: in a strong tape that can exceed 100%, in a deteriorating one it falls toward zero.

A momentum ETF gives you the stock selection and stays roughly fully invested through every environment, including 2008 and 2022. HELIOS uses a comparable style of selection but adds the exposure layer — sizing down as participation breaks and up when the book is calm. That cuts both ways: in a sharp V-shaped recovery, a dial that moved to cash re-deploys after the bounce has begun, trading some upside capture for a shallower path through bear markets.

No, deliberately. Exits are relative — a stock is sold when it falls out of the ranking at the next rebalance, not when it hits a price. A momentum book with price stops tends to sell into ordinary volatility and buy back higher. Downside protection instead comes from the exposure dials bleeding gross exposure toward cash as participation deteriorates and volatility rises — a dimmer, not a switch.

No. A leverage-proportional VIX hedge and systematic index shorts were both tested and rejected. Because the dials already de-lever into a deteriorating market, a hedge sized to leverage gets sold down exactly as a spike arrives — in the 2020 crash the tested hedge added nothing to drawdown reduction while paying roll decay through calm stretches. The volatility target and regime dial are the only risk layer.

You take it. The dials use prior-day data and react over days, not intraday, so a single-session shock hits the book at whatever exposure it carried into that session. The design protects against sustained declines — 2000–02 and 2008-style markets where participation erodes over weeks — not a one-day dislocation from a calm start.

Twenty names at 5% each, with no sector limits applied. Momentum clusters by nature — it's entirely normal for the book to hold eight or ten names from one leading sector. That's deliberate rather than accidental (capping it would mean overriding the ranking with a discretionary view) and it's a genuine risk to size your allocation around.

Yes, at the levered settings. Average gross exposure runs roughly 0.8×–1.3× depending on the cap you choose, peaking at the cap in healthy, low-volatility markets and falling near zero in risk-off periods. Leverage here is applied to hit a risk level, not to chase return — the volatility target sets it.

You can run the unlevered version. US retirement accounts — IRAs, Roth IRAs, most 401(k)s — can't use margin borrowing, so unlevered is the only setting available in those accounts. The upside: the tax friction described in Section D disappears entirely inside them.

That depends on your broker. Margin eligibility, maintenance requirements, commission schedules and financing rates vary widely across US brokerages, and the higher leverage settings generally require a portfolio margin account rather than standard Reg T. Check specific terms with your own broker before choosing a setting — those costs are yours, not ours, and aren't reflected in any figures we publish.

Not in practice. Rebalances are monthly and exposure adjustments are position-size changes rather than same-day round trips, so the strategy doesn't generate the day-trade pattern the PDT rule targets.

Mostly as short-term capital gains, taxed at ordinary income rates — up to 37% federal, plus the 3.8% NIIT where applicable, plus state tax. A roughly monthly holding period means few positions reach the one-year mark for long-term treatment, so any comparison against buy-and-hold — which defers gains indefinitely — should be made after tax, not before.

They can. If a position sold at a loss re-enters the portfolio within 30 days, the loss is disallowed under the wash sale rule and added to the new position's basis — names that oscillate around the ranking boundary are exactly the ones prone to this. It defers rather than destroys the deduction, but complicates year-end planning.

A tax-deferred or tax-free account, if you have the room — none of the short-term tax treatment above applies inside an IRA or Roth. The trade-off is those accounts can't use margin, limiting you to the unlevered setting.

Yes — reported figures are total returns, inclusive of dividends. In practice dividends are incidental to this strategy (momentum leaders often pay little or nothing), and a holding period under 61 days may fail the qualified-dividend test, taxing those dividends at ordinary rather than preferential rates.

The sleeve draws from sixteen funds spanning energy, precious and base metals, agriculture, uranium and carbon, holding whichever are in positive medium-term momentum at the time. Tax treatment isn't uniform: several futures-based funds are organised as partnerships issuing a Schedule K-1 rather than a 1099, which can delay filing, while the equity- and trust-structured funds on the list don't. Confirm the current treatment of each with your tax adviser rather than assuming they behave alike.

Yes. The strategy's live, time-stamped record is published on Collective2, the third-party platform through which it's executed. Because Collective2 records every signal as it's issued rather than after the fact, that record — not any simulation — is the one to judge HELIOS on. See the Collective2 strategy page ↗.

The test universe includes delisted stocks wherever such data exists, so companies that failed, merged or were taken private remain in the historical record rather than being quietly dropped from it. A universe built only from companies that still exist today implicitly knows which ones survived, and that hindsight inflates results — including the delisted names removes that advantage.

You cannot know it with certainty, and anyone claiming otherwise is overselling. In the methodology's favor: every overlay uses only prior-day data with no look-ahead, the universe is re-screened at each point in time, the parameter count is deliberately small, and components that failed — a resilience tilt, a dynamic exit rule, a volatility hedge, an always-on short sleeve — were removed rather than retuned. Against that: the parameters were still chosen with knowledge of market history, over a single historical path. The live record is the honest test.

Two distinct failures. The first is momentum ceasing to pay — a market that mean-reverts sharply, where yesterday's leaders systematically become tomorrow's laggards. The second is the regime dial misfiring — a decline so fast that participation never signals before the damage, or a series of false risk-off signals that hold the book in cash through a recovery. A narrow, mega-cap-led market is the known soft spot: when a handful of names carry the index, a twenty-stock equal-weight book tends to lag.

Daily closing prices only — for the candidate universe, for the broad universe that sets the regime reading, and for the commodity ETFs. No fundamentals, no analyst estimates, no alternative data, no intraday feed. The full input set is public end-of-day pricing.

No. Execution is fully AutoTraded through Collective2, a third-party trading platform — signals are transmitted to the platform, which routes the corresponding orders to your connected brokerage account without you acting on each one. Your account, your broker and your capital remain entirely yours throughout; Collective2 sits between the signal and your broker as the execution layer. Collective2 lists which brokers currently support AutoTrade on the strategy page — check that yours is included before subscribing.

Very little day to day, since execution is automated. Your decisions are the ones made up front: which leverage setting to run, how much capital to allocate, and in which type of account. After that, monthly rebalances and daily exposure adjustments happen without intervention. What still requires your attention is oversight — checking the AutoTrade connection is live, your account has the buying power the setting assumes, and fills are tracking the published signals.

It self-corrects. The strategy is stateless — each run recomputes the entire target portfolio from full history rather than adjusting yesterday's positions, so a missed or partially filled rebalance doesn't corrupt anything downstream; the next run states the correct portfolio and the gap closes. It does mean you carried different positions, and their returns, in the interval — sustained tracking differences are worth investigating rather than ignoring.