Long UPRO shares instead of SPY calls for a 2-week directional view — critique my structure choice

Posting this for critique, not validation. I want to be told where the structure is wrong before the market tells me.

**The position** 4 shares UPRO @ 99.16, currently 108.00, +8.91% unrealized. \~$432 notional, \~$1,300 equivalent SPX delta-1 exposure. Hard stop at 105. 2-week horizon.

Small capital, I know. The question isn't the size. The question is whether the structure is the right one for the view.

**The view**

Directional long SPX, 2-week horizon, driven by the AI capex tape refusing to roll over on heavy headline flow. I wanted leveraged exposure without paying vol premium for a move I couldn't time to the day. That pushed me toward shares of a 3x LETF instead of calls.

**The structures I compared before entering**

*SPY 2-week ATM calls (\~25 DTE, 30 delta)*: roughly $3-4 premium. Theta \~$0.15/day. Break-even needs about a 1.5% SPX move just to cover premium and decay. Upside convexity is real, a 5% rip makes these print 3-4x. But I'm short theta the whole way and if SPX chops for five sessions I bleed while I'm directionally right.

*SPY debit spread (560/570 2-week)*: cleaner theta profile, caps upside, reduces IV risk. Fine structure, but at $432 of capital I'm buying 2-3 spreads and the max profit is bounded. Felt like I was paying for structure I didn't need at this size.

*UPRO shares (what I did)*: delta-1, roughly 3x SPX delta per share. No position-level theta, no IV crush, no roll decision. The only decay is the LETF internal volatility drag, which at current realized vol (\~10-12%) compounds in the low tens of bps over 2 weeks. Over 6 months it's brutal. I'm not holding 6 months.

*UPRO calls*: looked at them, spreads were garbage, bailed.

*Short 2-week OTM SPY put*: defined view, undefined risk at this account size, not doing it.

**The tradeoff I'm explicitly accepting**

I'm giving up convexity. A 5% SPX rip makes SPY calls print multiples; UPRO prints \~15% which on $432 is $65. The calls would be the better expression if I knew the move was this week. I don't. I know the structure is intact and I think the move comes inside two weeks. For a view that's "directional up with uncertain timing," shares of the LETF price better on risk-adjusted terms than short-dated calls, because I'm not fighting the clock.

**Where the decay math breaks my thesis**

The volatility drag is a function of realized vol, not implied. At SPX realized \~10-12% the 2-week drag is small enough to ignore. If VIX pops from 14 to 25 and we chop instead of trend, the daily-reset mechanics eat me even if SPX ends flat. That's the real structural risk and it's not priced into my $12 dollar-risk number, which assumes directional stop-out, not decay stop-out.

The signal I'm watching: VIX term structure. If front-month crosses M2 into backwardation, the decay regime changes and I'm out regardless of price. That's a harder exit rule to execute than a price stop but it's the one that actually protects the structure choice.

**Risk at position level**

Stop at 105, \~2.8% below spot, \~1% SPX through the line given 3x. Max loss from here \~$12, from cost basis \~$15. Unrealized $35, so from here it's effectively a free roll. I'm trailing behind the 2-hour structure once it tags 110 and I'm out immediately on any daily close below the stop. Not moving the stop, not averaging down, not negotiating with the tape.

**Three specific questions**

1. For a defined 2-week directional view at sub-$500 capital, does anyone have real back-test data on LETF shares vs. short-dated ATM calls on risk-adjusted basis? My prior is that LETF shares win under \~3 weeks because you're not paying the vol premium, but I've never seen it rigorously tested and I'd like to.
2. How are you actually monitoring LETF decay intraday on a live position? I have the formula for expected drag but I don't have a clean way to see whether realized vol is running hotter than my assumption without pulling data after the close.
3. For the VIX-backwardation exit signal, anyone have experience on how fast it develops? I'm worried about regimes where the inversion happens in a single session and by the time I see it on my screen the damage is already in the LETF NAV.

Not asking for a target, not asking where SPX goes. Asking whether the structure choice is defensible for the view, and what I'm missing.

Edit: spy as of 12;12 est is @ 680.30, bulls are running