I am developing an algorithm and it needs to know what to do in certain market conditions
It takes on a Vertical Bull Call Debit Spread on LEAPS that are 12+ months out in the future. This means that a long OTM call is taken, and a short OTM call at a higher strike price is acquired as well. The net debit will usually be small, like .20 or .50 or 2.00 , this allows it to take on MANY contracts with very little capital
It gives only minuscule downside protection (to the entire position I mean, the entire position can take a loss quickly but it is nothing in compared to if you were only long that many contracts), but you also pick up theta decay from the short options if things don't move fast enough in your time frame
My problem is legging out, because it looks like too much leverage with my calculations.
The trading plan:
Buy long OTM calls, sell short further OTM calls
In the future, 6 months in, after some theta had some time to burn , if still bullish - close the short leg for 50-80% gain
Use the proceeds from closing the short leg to average down on the long leg (the long leg will be loss making but acquirable for much cheaper, this gives you control over a lot more contracts)
On the first uptick you make 50% gain (on the original principal used to open vertical position), on real bullish moves you make 4000% gains
- If market direction turns against you, still max 100% loss on original position. Capped loss. May mitigate this with a stop loss, but stops may be too tight when you have 4 months to go.
The "problem" is how much leverage this appears to be. I am looking at 4000% to 10000% gain on original principle. Simple put, this many contracts would not have been affordable without collecting the premium from the short options. Yet the risk is still sound and limited to 100%, especially with any position sizing done in a larger portfolio. So I'm trying to find what I'm missing. This is a bullish strategy and it appears worthwhile to periodically buy small amounts of put protection because it will keep your losses near 0% when it is just a vertical.
I have my backtest going that shows everything I just described (which is really what makes me skeptical), so I would like the community to illuminate some things about verticals that I haven't already considered