Though TNA & TZA are hedges against each other; you dont want to put ic's in both. Think.. the worst case scenario for either of the ics is a big movement. And when that happens in one it happens in the other just in the opposite direction. So you will get double whammied. Just pick one; its doubling up. Just pick one -and you can skew it to express your directional bias one way or the other; or the lack thereof. -gariki
I haven't analyzed it, but I would think something like long ATM puts in two anti-correlated 3X ETFs would have some potential. Gamma helps you on each side. A big move is capped in one but uncapped (until underlying goes to 0) on the other. I guess it's all in the price.
Newbie at this forum. Apologies if I'm treading on old toes. I blogged about @jimmyjazz's thought above. He's right - it does come down to price. Most 3x levered ETFs are HTB, so put/call parity doesn't apply and premium on the put side is ridiculous. Going on the call side for 3x levered ETFs means you've vol-drag (and usually futures contango) going against you. A compromise is, say, buying puts on something like TZA (vol-drag and contango on your side) and hedging with RUT put options.
But If you're a credit seller on them and have long term plays, which have ups and downs, can't you benefit on the swings and the theta decay over time? especially since they will both be rich in premium? Thats assuming that the premium received is enough to cover losses over a large number of occurences. Another thought I had, which I assume most people already know is - Since the market is pretty much random and normally distributed, i.e for every occurrence of a down move there is a near equivalent number up moves, can't we predict when up trends start to form? When looking at the charts, it seems every wave is pretty much a mini bell curve. The only difference is that the magnitude of up and down moves may not be equal. with that information, can't you plan high probability setups? i.e buying on down moves of 1.5 to 2 std?