Leverage as an accelerant, not a default

Leveraged ETFs get discussed as if they were one thing. They are not. The same ticker can be a reasonable tool inside one plan and a slow leak inside another, and the difference is almost entirely in how it gets used. There are two broad patterns. The first treats leverage as a standing default: pick a 2x or 3x fund, hold it the way you would hold an index fund, let the multiplier do the work. The second treats leverage as an accelerant: a deliberate substitution, made only on the highest-conviction setups, at reduced size, for a bounded stretch of time. This post is about why the second pattern is the defensible one, and how Coil (coil.trade) applies it.

What the standing default actually signs you up for

A 2x or 3x ETF targets a multiple of its index's daily return. Not its yearly return, not its return since you bought. Its daily return. That detail changes everything about holding one for a long time, because daily multiples compound. In a smooth, persistent trend the compounding can help. In a choppy, sideways market it works against you: the fund can bleed value while the index it tracks finishes flat. That behavior goes by volatility drag or decay, and the mechanics are covered in the leveraged ETF decay explainer.

Buy-and-hold leverage is therefore not really a bet that the market goes up. It's a bet on the path the market takes on the way up. Two futures with the same endpoint can treat a 3x holder very differently depending on how rough the trip was, and in the futures that include a deep index drawdown, the leveraged version of that drawdown can get severe enough that recovery math stops being realistic. Add overnight gaps, which no intraday exit can protect against, and the standing default looks like what it is: an always-on amplifier pointed at whatever happens next. There's more on that failure mode in the gap risk post.

Leverage as an accelerant

The alternative isn't to swear off leverage. It's to demote it. In the accelerant model, leverage is never a portfolio stance. It's a per-trade decision, made after everything else about the trade has already been decided. You've picked the name. You've chosen the entry. You've placed the stop. The last question is which vehicle expresses the trade, and on a small minority of setups, the ones where conviction is highest, the answer can be a leveraged version of the underlying instead of the underlying itself.

Framed that way, the daily-reset problem mostly dissolves. Decay is a function of time and chop, and a position held for days or weeks with a defined exit doesn't give the drag much room to accumulate. What matters instead is that the trade itself is good: right name, right spot, honest invalidation. Leverage doesn't fix a mediocre trade. It makes a mediocre trade worse, faster.

Sizing is what keeps risk constant

Here's the part most discussions skip. When a 2x vehicle substitutes for the underlying, the position size should drop with it, roughly in half. A half-sized position in a 2x fund carries about the same first-order exposure as a full-sized position in the plain one. The daily dollar swings look similar. The risk budget for that name hasn't grown. What changed is the shape of the position: less capital committed for the same exposure, and a hard cap on the absolute dollars at stake in the worst case, because a fund can only take what you put into it.

The single-stock case makes the trade-offs concrete. NVDL against NVDA, for instance, is a different instrument with its own costs and path behavior, which is exactly why it should be a considered substitution and never a reflex. That comparison is written up in NVDL vs NVDA.

Conviction and structure both have to agree

So when does the substitution earn its place? Two independent tests, and both have to pass.

Conviction is the first. The name has to sit at the top of whatever ranking process you actually trust. A genuine leader, not a beaten-down ticker you hope will bounce. If the setup is merely fine, the leveraged version isn't justified. Take the plain trade, or no trade.

Structure is the second. The entry has to be at a real level: a pullback to support that has held before, with a stop placed where the thesis is objectively wrong rather than where the pain gets uncomfortable. That's the difference between a structural stop and an emotional one, covered in structural stops explained. Structure matters twice as much under leverage, because every mistake is amplified. A sloppy entry that costs a little in the underlying costs double in the 2x wrapper.

Conviction without structure is enthusiasm. Structure without conviction is a setup that doesn't deserve the amplifier. The accelerant is reserved for the intersection.

This is how Coil treats it

Coil runs this model literally. Its scanner scores every S&P 500, Nasdaq-100 and Macro-book name on leadership and entry quality, and its engine trades the published scores by rule: it buys leaders pulling back to real support, never chases, sizes by conviction, and goes to cash when nothing qualifies. On a small set of high-conviction names it may substitute a leveraged ETF, NVDL or SPXL for example, long side only and at reduced size. It never holds leverage as a standing default, never shorts, and never touches inverse ETFs. The research behind the ranking is laid out at /how-it-works.

None of this makes leverage safe. It makes leverage accountable: used on purpose, in a defined spot, with the size cut that keeps the risk budget honest, and dropped the moment the setup that justified it is gone.

FAQ

Is a leveraged ETF at reduced size the same risk as the underlying at full size?

To a first order the daily exposure is similar, which is the point of cutting size when you substitute. It is not identical: daily resets, fund costs and overnight gaps make the leveraged wrapper behave differently over time, and it can lose value much faster in a crash. The smaller position caps your worst case in dollars, but nothing removes the risk.

Why not just buy and hold a 2x or 3x fund instead?

Because those funds compound daily returns, long holding periods turn a market view into a bet on the market's path. Sideways chop erodes them even when the index goes nowhere, and a deep index drawdown gets far deeper at 2x or 3x. Short, deliberate holding periods on strong setups avoid most of that. Permanent holding absorbs all of it.

Does Coil ever use leverage on the short side?

No. Coil is long-only. It never shorts and never uses inverse ETFs. When nothing qualifies under its rules, it holds cash rather than manufacturing a position.