At 2 a.m., I pulled up the GT I opened last week on TermMax and stared at it for ten minutes. $ETH was clearly rebounding, yet its quote kept drifting downward—within the same day it sank another chunk. I still remember this scene to this day.

Back then, there were only two days left until expiry. At first I couldn’t make sense of it. The leveraged position tracks the underlying—if the underlying is going up, why would the GT bleed lower? Only after I took the GT apart did I understand: its debt leg is the FT, a zero-coupon bond that must normalize to par at maturity. The closer the date, the more the FT’s discount converges toward zero. The identity P(FT) + P(XT) ≈ 1 is set in stone: as the FT climbs, XT must fall. After stripping out the debt, the remaining exposure in the GT keeps leaking—what it leaks has nothing to do with direction, it’s simply time collecting rent.

At the time, I hadn’t fully distinguished between ordinary GT and the option-style structure of Alpha yet. I only knew that the regular GT position also has a LLTV liquidation line; if the price breaks through, it triggers a close. This near-expiry GT that kept grinding lower taught me that lesson first.

That incident made me rethink the fundamental difference between TermMax’s GT and Perp. Perp has no endpoint. Paying funding is like paying to keep renting the exposure—the cost drifts with the funding rate. GT, on the other hand, has a fixed endpoint. Pre-paying the option premium buys out the risk for a fixed term. Even if it doesn’t “settle” before expiry, it still must settle at maturity; the cost is locked in the moment you open the position. If you want to keep holding, you roll into the next expiry and pay the premium again. And because TermMax isolates markets by term, when you roll the interest-rate curve you encounter is redrawn—your cost isn’t necessarily the same as the previous round.

So that last time I took my 2x near-expiry position and exited. It wasn’t that I was wrong about direction—I just worked out the accounting that rolling requires paying the front-end interest again, and found that the closer you get to expiry, the less worthwhile it is to renew the same exposure for another round of time decay. After that, before I touch a GT, I first wait and watch TermMax’s expiry distribution and only move when I pick a term farther from settlement; the leverage multiplier is secondary. #termmax @TermMax