A tactical long-volatility position, opened small and built in tranches. The logic is asymmetry: when the tape is this extended and complacency this high, the cost of being early is small and the payoff to a pullback is large.
Every long-volatility trade is really a bet about complacency, and by late May the tape was about as complacent as it gets. The S&P 500 had closed green for nine consecutive weeks. That sounds unremarkable until you look at the base rate: across the last three decades the index has only managed ten green weeks in a row on three occasions. Streaks like this do not end because some indicator flips — they end because the marginal buyer runs out, and the unwind, when it comes, is fast.
Alongside that, price was pinned at standard-deviation extremes on both the 6-month and 1-year channels. Stretched tape, low realised volatility, a VIX grinding near its lows for the year — the ingredients for a sharp mean-reversion in volatility were all present. None of that tells you the day it turns. It tells you the odds have shifted, and that owning convexity is cheap relative to the payoff if it does.
VXX gives exposure to short-dated VIX futures, so it is the cleanest liquid way to be long expected volatility without running an options book. But it comes with a tax: the VIX futures curve is usually in contango, so the ETN bleeds a little every day it is held in calm markets. That single fact dictates everything about how the position is run.
You do not buy VXX and wait. You size it small, you have a plan to add only on strength in your favour, and you take profit aggressively when volatility spikes — because the same convexity that makes the spike violent also makes it fleeting. This is a trade you are renting, not owning.
The position opened at just 2.5% of the portfolio. That is deliberate: if the timing is early — and with volatility trades it often is — the cost of being early stays trivial, and there is dry powder to add. The plan from there is to build in tranches as the VIX falls from around 15 toward 11, roughly 20% of the intended size at each step, to a maximum of 12.5% of the portfolio. Adding into lower volatility improves the entry and respects the contango drag.
When the move came, the position was trimmed hard — 80% sold into the spike at 25.08 — leaving a smaller runner. That is the discipline a decaying instrument demands: bank the convexity when the market hands it to you rather than waiting for the perfect exit that contango will quietly erode. The remaining runner was sold at 27.00 — into the 6-month standard-deviation mean and the round 27 psychological level — closing the trade for a clean win, a touch shy of the 27.20 target. A textbook exit for a decaying instrument: take what volatility hands you rather than waiting for the last tick.
The clearest risk is simply that the melt-up continues. A stretched tape can get more stretched, and every week that passes without a pullback costs a little carry. The instrument itself is the second risk — VXX is built to decay, so patience is expensive in a way it is not for an equity.
The thesis is invalidated by a tape that keeps making new highs on steady, low realised volatility while the VIX grinds lower — at which point the carry cost outweighs the diminishing odds of a near-term shock, and the runner is cut. This is a probabilistic, time-boxed trade, not a conviction hold.
The live position. This idea is tracked with real entries, exits and option legs on the Trade Log (VXX). Educational only — not financial advice or a recommendation.