Many people perceive 7709 as: when SK Hynix rises 1%, 7709 rises 2%.

It sounds incredibly simple.

Many even believe it's safer than perpetual contracts because it lacks an explicit liquidation level and won't suddenly "blow up" like a futures account.

But once you truly understand its daily rebalancing mechanism, you'll realize: 7709 is not simply "two times Hynix."

It is essentially a leverage strategy that operates long-term, mechanically chases rallies and sells off declines, continuously bears volatility decay, and also pays financing and derivative costs.

It bears the negative path effect similar to an option seller's Short Gamma, yet receives no IV and Theta compensation that an option seller deserves.

This is the core reason why I believe it is inherently negative EV.

1. Why Must 7709 Rebalance Constantly?

Assume the fund's net asset value (NAV) is 100, and the target leverage is 2x. Then it needs to maintain a 200 exposure to SK Hynix.

If Hynix rises 10%, the fund makes roughly 20, and the NAV becomes 120.

The original 200 position, after the rise, has a market value of 220.

The actual leverage now becomes:

220 ÷ 120 = 1.83x.

To restore it to 2x, the fund's target exposure should be:

120 × 2 = 240.

Therefore, it must add 20 of exposure *after* Hynix has already risen.

Conversely, if Hynix falls, the fund's NAV declines faster than the market value of its holdings, causing the actual leverage to exceed 2x.

To bring the leverage back down to 2x, it must sell part of its position *after* Hynix has already fallen.

Thus, the rebalancing direction of such leveraged ETFs is always:

Buy after rising; Sell after falling.

In other words, standard: chasing rallies and selling off declines.

2. Why Do I Say Each Rebalancing Loses Money?

The "losing money" referred to here doesn't mean the order shows an immediate loss upon execution.

It means:

Each rebalancing is a delayed reaction to already occurred price movements.

Hynix has already risen from 100 to 110, then the fund realizes its leverage is insufficient and adds exposure around 110.

But the appropriate time to increase exposure was before the rise, or at least during the rising process.

Now, buying after the price has already risen a segment inevitably means a higher purchase price.

Similarly, Hynix has already fallen from 100 to 90, then the fund realizes its leverage is too high and reduces exposure around 90.

The appropriate time to reduce the position was before the fall, or at least during the falling process.

Now, selling after the price has already fallen a segment inevitably means a lower selling price.

So, from the perspective of the goal of "continuously maintaining 2x leverage," there's a certain fact for each rebalancing:

It's always one step behind.

When rising, it earns less than the profit it could have gained by increasing exposure earlier;

When falling, it suffers more loss than it could have avoided by reducing exposure earlier.

This is the loss caused by delayed rebalancing.

Of course, some may argue:

Adding after a rise—if the price continues to rise afterward, won't the newly added position still make money?

Of course it can.

But that's profit generated by the next segment of price movement.

After subsequent price movements occur, the fund will rebalance its position again based on the new rises or falls.

You cannot use the possibility of the next segment continuing to rise to deny the opportunity loss that has already occurred in the previous segment because the fund didn't maintain 2x exposure promptly.

More accurately:

Past price movements determine *why* to rebalance and *how much*;

The current rebalancing resets the future risk exposure;

Future price movements determine the actual profit or loss after the new position is added.

However, compared to an ideal strategy that can maintain 2x leverage in real-time, the loss from delayed rebalancing has already occurred.

3. Even in a One-Way Uptrend, Daily Rebalancing Has "Delayed Return"

This point is especially important.

Some might think that as long as Hynix keeps rising, 7709 can achieve good compounding by continuously adding after rises, so rebalancing isn't a problem.

But in reality, during a one-way uptrend: the higher the rebalancing frequency, the higher the return.

Because profits can be reinvested earlier.

Take a simple example.

Assume Hynix's price movement for a day is divided into two stages, each with a consecutive 5% rise.

Hynix's cumulative daily rise:

1.05 × 1.05 - 1 = 10.25%.

Assume 7709 has a morning NAV of 100, initial exposure of 200, and doesn't rebalance all day, only restoring 2x leverage near the close.

Then the day's return is approximately:

10.25% × 2 = 20.5%.

NAV changes from 100 to 120.5.

But if it immediately restores leverage to 2x after the first 5% rise in the first stage, and then participates in the second stage rise, the result becomes:

First stage: Hynix rises 5%, 2x product rises 10%:

100 becomes 110.

The fund immediately restores exposure to 2x.

Second stage: Hynix rises another 5%, 2x product rises another 10%:

110 becomes 121. Final return is 21%.

For the same one-way uptrend: rebalancing once a day yields 20.5%;

Rebalancing once mid-day yields 21%.

If changed to rebalancing every half hour, profits can be reinvested even earlier, and the final return would further approach the theoretical result of continuously maintaining 2x leverage.

This illustrates: even in a one-way uptrend, rebalancing only once a day also suffers from delayed return.

The fund didn't fail to earn 2x return; rather, it reinvested profits too late, thus failing to earn the more complete 2x compounded return.

The same applies to a one-way downtrend.

The more timely the rebalancing, the sooner the fund can reduce its position, thereby lessening subsequent losses.

Rebalancing only near the close each day essentially involves double delay:

During uptrends, not reinvesting profits promptly, thus earning less;

During downtrends, not reducing excessive leverage promptly, thus losing more.

Therefore, if rebalancing frequency is changed from once a day to every half hour, the result would be better in one-way trends.

Only continuous rebalancing would most closely approximate true constant 2x leverage.

4. But Higher Rebalancing Frequency Aggravates Decay in Choppy Markets

The problem is, higher frequency rebalancing is not free either.

Assume Hynix rises from 100 to 110 intraday, then falls back to 100.

Ultimately, Hynix has no net change.

If 7709 does not rebalance all day and calculates only at the close, theoretically its NAV change is close to zero.

But if it restores 2x leverage at the 110 level, and then Hynix falls from 110 back to 100, it will absorb the subsequent decline with a larger position.

First segment: rise 10%, NAV from 100 to 120.

After restoring 2x leverage around 110, second segment: fall from 110 to 100, a drop of approximately 9.09%.

The 2x product's NAV then becomes:

120 × (1 - 18.18%) ≈ 98.18.

Hynix ends back at the starting point, but the product loses about 1.82%.

This demonstrates an inherent, unresolvable contradiction for leveraged ETFs:

Low rebalancing frequency leads to more severe delayed return and delayed loss-cutting in trend markets;

High rebalancing frequency leads to more frequent chasing of rallies and selling off declines in choppy markets, generating more severe volatility decay, spreads, slippage, and transaction costs.

Therefore, rebalancing only once a day doesn't eliminate this structural issue.

It merely makes a choice between: delayed loss in trend markets and frequent rebalancing loss in choppy markets.

5. Why Is It Like an Option Seller's Dynamic Hedging?

Those familiar with options can easily grasp this logic.

Assume a trader sells a large number of Calls and Puts and constantly performs Delta hedging.

Because option sellers are typically Short Gamma:

When the underlying rises, the portfolio's Delta increases, and the trader needs to buy the underlying;

When the underlying falls, the portfolio's Delta decreases, and the trader needs to sell the underlying.

Thus, the option seller's Delta hedging is also:

Buy after rising; Sell after falling. Also chasing rallies and selling off declines.

When the market constantly oscillates up and down, the seller-trader continuously buys high and sells low. The greater the realized volatility, the larger the Gamma loss typically generated by dynamic hedging.

But why can option sellers still make money?

Because they receive the option premium upfront.

The premium contains:

IV, or implied volatility;

Theta, or time value.

As long as ultimately: the received IV and Theta exceed

the realized volatility loss, jump loss, and transaction costs that occur,

the seller's strategy can be profitable.

In other words, an option seller bearing Short Gamma is not doing so for free.

The market pays them a volatility risk premium as compensation.

6. The Biggest Problem with 7709: It Bears Similar Short Gamma Decay but Receives No Theta

7709 also:

Adds after rising; Reduces after falling;

Repeatedly chases rallies and sells off declines in choppiness;

The higher the realized volatility, the more severe the path decay.

So, from the perspective of trade flow and path dependence, it exhibits characteristics similar to negative Gamma, negative realized variance.

But investors in 7709 do not receive an option premium.

No one pays you an IV upfront for bearing this path risk of chasing rallies and selling off declines.

It also lacks a definite expiration date; there's no Theta that gradually accrues to you over time.

It is a mechanized rebalancing strategy that approximates perpetual operation.

Moreover, not only does it not receive volatility premium, it must continuously pay: Swap financing costs, option costs, management fees, bid-ask spreads, rebalancing slippage, market impact, currency exchange and other product fees, secondary market premium/discount convergence risk.

Thus, its true structure is closer to: two times directional return minus realized variance drag minus financing cost minus derivative cost minus management fee minus transaction cost.

This is why I believe it is inherently negative EV.

7. What Exactly Does "Negative EV" Mean Here?

I'm not saying: no matter how much Hynix rises, 7709 will definitely lose money.

If Hynix experiences a very strong, very persistent, very smooth one-way uptrend, directional returns could certainly cover all decay, and 7709 could also make significant money.

But this doesn't mean the product structure itself is positive EV.

Games in a casino can also have winners, but that doesn't change the fact that the rules are long-term disadvantageous to players.

The negative EV I refer to means:

In obtaining the same directional exposure, 7709, compared to an ideal 2x strategy, low-cost perpetuals, or investors managing leverage themselves, has an additional layer of deterministic negative carry, path decay, and product fees.

Investors must not only correctly judge that Hynix will rise, but the magnitude of the rise must be sufficiently large, the trend sufficiently persistent, and the price path sufficiently smooth to cover the costs continuously deducted within the product.

In other words: investors need to be right about the direction *and* the path. Being right about the final price alone is not enough.

For example, Hynix ultimately rising 30% doesn't guarantee 7709 will achieve a 60% return.

If the intermediate path experienced multiple: sharp rises, sharp falls, rebounds, and declines again; even if Hynix ultimately returns to a high level, 7709 might significantly underperform a simple 2x cumulative return due to daily leverage reset and volatility decay.

8. Why Are High Volatility and Frequent Reversals Most Detrimental?

Assume Hynix: rises 10% on day one; falls 9.09% on day two.

Hynix ends back at the starting point. But a 2x daily leverage product:

Rises 20% on day one, NAV from 100 to 120;

Falls approximately 18.18% on day two, NAV from 120 to about 98.18.

Hynix ultimately doesn't lose money, but the leveraged ETF loses about 1.82%.

If such oscillations continue, the NAV will be continuously eroded.

Every market reversal punishes its previous rebalancing.

So what these products truly fear is not just one-way declines, but: high volatility,剧烈震荡, frequent reversals; mean reversion.

And SK Hynix itself is a highly volatile single stock.

Placing a daily leverage reset mechanism on such an underlying typically results in more severe long-term path decay compared to low-volatility index-based leveraged ETFs.

9. "No Liquidation" Is Just Packaging; It Doesn't Mean NAV Can't Approach Zero

Many investors like leveraged ETFs because they appear not to trigger sudden liquidation at a specific price like perpetual contracts.

But this merely transforms explicit liquidation into implicit NAV decay.

During continuous declines, the fund continuously reduces its position:

Lower NAV, smaller position;

Smaller position, smaller absolute loss next time.

So in many continuous decline paths, it won't end suddenly at a liquidation line like a personal futures account.

However, its NAV can continuously approach zero.

From 100 to 20 is already an 80% loss.

At this point, it's not that an 80% rise recovers the loss, but a 400% rise is required.

Therefore, the so-called "no liquidation" often just means:

There's no clear liquidation moment;

But principal can still approach zero amidst continuous chasing of rallies and selling off declines, volatility decay, and fee erosion.

It transforms liquidation into a slow process.

10. Why Changing from Fixed 2x to "Up to 2x" May Not Benefit Existing Investors?

Such products later changed from a fixed 2x to an elastic leverage structure of "not exceeding 2x."

From a risk management perspective, this can reduce: daily rebalancing scale, Swap capacity requirements, market impact, and NAV collapse speed in extreme scenarios.

After leverage drops from 2x to 1.2x, 1.1x, the required chasing of rallies and selling off declines transactions decrease significantly.

But for existing investors deeply underwater, this may create an awkward structure:

During the decline phase, bear losses with near 2x leverage;

After severe market volatility, to control risk, the product lowers leverage to 1.1x or 1.2x;

Subsequently, even if Hynix rebounds, investors can only participate in the recovery with lower leverage.

In other words: high leverage on the way down, low leverage on the way up.

This doesn't necessarily imply malicious intent by the manager.

Reducing leverage indeed decreases future downside risk.

But from a product structure perspective, it also reduces the ability of already deeply underwater investors to recover quickly.

And the fund can still continuously charge management fees, and derivative counterparties can still continuously earn financing, Swap, or option-related income.

11. Why Might Perpetual Contracts Be More Efficient Than 7709?

Assume a Hynix perpetual contract exists in the market that is: sufficiently liquid, has reasonable Funding, reliable oracles, no severe premium/discount, and transparent risk management.

At a minimum, professional investors can decide for themselves:

Whether to maintain a fixed position

Whether to restore 2x leverage daily

Whether to rebalance every half hour

Whether to increase rebalancing frequency when trends are evident

Whether to reduce leverage in choppy environments

Whether to cut losses early

Whether to add margin

Whether to use other assets for portfolio hedging

Whereas 7709 encodes all these decisions into product rules.

Investors can only passively accept:

When to rebalance

How much to rebalance

What the actual target leverage is

How much Swap is used

How much option is used

How much derivative cost is borne

When leverage is reduced

Of course, perpetuals also have risks: Funding, liquidation, trading platform credit, oracles, liquidity, ADL, etc.

So perpetuals are not a free lunch.

But for those who truly understand managing margin and positions, they can at least control themselves: rebalancing frequency, leverage level, stop-loss rules, financing cost, holding period.

Rather than handing all decisions over to a set of mechanized product rules.

12. Final Summary

7709 is not simply "two times Hynix."

It is more like: two times long Beta plus mechanized rebalancing similar to Short Gamma

minus realized variance

minus financing cost

minus derivative cost

minus management fee

minus transaction cost

Each of its rebalancings is a delayed reaction to already occurred price movements:

Adding after rising;

Reducing after falling.

In one-way trends, if it rebalanced every half hour, it would reinvest profits earlier and reduce downside exposure earlier than rebalancing only once a day, resulting in better performance.

But in choppy markets, more frequent rebalancing leads to more instances of buying high and selling low.

Thus, it suffers from an inherent, inescapable structural dilemma:

Rebalance slowly, suffer delayed loss;

Rebalance quickly, suffer choppiness decay.

More importantly, it bears the chasing of rallies and selling off declines similar to an option seller's dynamic hedging, yet receives none of the IV and Theta compensation an option seller rightfully deserves.

The only compensation it can rely on is Hynix's future sufficiently strong, sufficiently persistent, sufficiently smooth uptrend.

Therefore, it is not a product that "loses money every single day."

But it is a product that inherently carries negative carry, negative variance, delayed rebalancing, and high holding costs—a negative EV product.

Investors need to correctly judge not only the direction, but also: trend persistence, volatility, price path, rebalancing frequency, financing cost, Swap and option costs, secondary market premium/discount.

For ordinary investors, even correctly judging the direction long-term is difficult, let alone judging so many variables simultaneously.

The so-called "no liquidation" merely hides liquidation risk within the process of continuously decaying NAV.

Appearing safer doesn't mean the structure is more advantageous.

Being right about the direction also doesn't mean you should choose the tool with the highest cost, worst path dependency, and lowest control.