Critique my strategy

Chuck Krug

Chuck Krug

Open an account A: Buy 100 shares of SPY
Sell 1 at the money call for a +-1.5% yield. Duration 30 days.

Open an account B: Short 100 shares of SPY
Sell 1 at the money put for a +- 1.5% yield with a 30 day duration.

Purpose: Harvest the time value of the options. Mimic an iron condor without having the theta decay of the long options.

Costs: Transation costs: a few bucks here and there at Interactive. Not that important
imo.
Borrow costs: 0.45% at Interactive Brokers.

Scenarios: SPY goes up 10-20% a year: You will have to from time to time. So you need a bit of extra cash to buy the shares back in account A. In account B the puts expire wortless.
Spy goes down 10-20% a year: vice versa.
 
Last edited:
Your net position in SPY is flat and you are short the at the money straddle, when you look at both accounts together. You just get to pay extra commissions in your scenario.

Good spread if the market doesn't move much, lots of risk if it does.
 
Suicide. Any gains are frozen, while all (market) risk -- on both sides -- remains.

If the market climbs, A=$0, while B's negativity grows in proportion.
If the market declines, B=$0, while A's negativity grows.

The bet you're laying is that the unlimited *bi-directional* liability to which you expose your account will be less than the ~3% revenue freed up upon expiration.

So..... reward=3% // risk=∞
 
tommcginnis said:
Suicide. Any gains are frozen, while all (market) risk -- on both sides -- remains.

If the market climbs, A=$0, while B's negativity grows in proportion.
If the market declines, B=$0, while A's negativity grows.

The bet you're laying is that the unlimited *bi-directional* liability to which you expose your account will be less than the ~3% revenue freed up upon expiration.

So..... reward=3% // risk=∞
More...

And to follow up on tommcginnis' 3% reward / infinite risk. You would have the same risk / reward by simply doing the short straddle and not messing about with the underlying.

Ultimately your total profit/loss is the combination of your accounts. Simple algebra should tell you your total profit/loss doesn't depend on the underlying.

Let position A1 be the options position in account A.
Let position A2 be the underlying position in account A.
Let position B1 be the options position in account B.
Let position B2 be the underlying position in account B.
PL is a function that represents the profit/loss of a position.

Total profit/loss = PL(position A1) + PL(position A2) + PL(position B1) + PL(position B2)

In your case PL(position A2) = -PL(position B2). Therefore, you have:
Total profit/loss = PL(position A1) - PL(position B2) + PL(position B1) + PL(position B2)
Total profit/loss = PL(position A1) + PL(position B1)

This is not the first time I've seen people suggest strategies across multiple accounts as though they've found some amazing idea when they ultimately all boil down to the maths above.

Simply put, the only way such a multi-account strategy would work is if you combine it an O'Hare spread.
 
It's just a short straddle like @FSU says. I don't know why you would need the 2 accounts... and you're paying more in fees.

Short straddle is short volatility... so you don't want any move.... or at least not beyond the straddle value.
 
Decades back, some bad CTA's would do something like this and when it is time to raise some money, they pull out the statements from the winning side and bury the story of the bad side..I am in no was suggesting that this is the case here.. just saying...you are syn short put in acct 1 and syn short call in 2 ... margining in account 2 might be diff than acct1 due to infinite risk.
 
Each of your strategies is a synthetic naked option explained via the Put call parity relationship.
Long stock plus short call equals short put.
You have unlimited losses outside of a 3% break even move.
So if spy is going up 20% is going to have a 17% loss and the same moving the other direction. Please trade small...much to learn.
 
There is a way to do it. A2 + B2 generates a profit. Plus A1 + B1 premium collection. Pick a statistical outlier as the strike price for A1 + B1, collect the premium. The price of A2 and B2 are slightly different, every time A2 or B2 goes into a loss, exit A2 or B2, keep the winning A2 or B2, since its a statistical outlier, you are only left with A2 or B2 in the end over time. The main thing is you have to be precise, in the 'statistical outlier'.. if your not, the costs of the transactions in A2 or B2 will exceed the premium collected from A1 and B1. The premium collected buffer's the transaction costs. Otherwise you could just trade the underlying only with same logic. But the transaction costs would mount to a point where the deviation from entry would have to be substantial for it to generate profit.

In summary your selling the straddle to pay for the transaction costs, aiming to keep transaction costs minimal. I believe this is what the originator of the thread was aiming for.

The question becomes, how do you predict when and where the statistical outlier will be. The drivers of volatility, create statistical outliers. Because of what has been mentioned above, AI will reduce the volatility, over time price action will just be a step pattern.. from one consolidation point to the next. Predicting the abrupt jump to new equilibrium is where AI/Machine learning will try to become expert at. AI will reduce Option prices over time, or arbitrage out the excess premium. Is that why volatility is decreasing as the years pass? Only niche markets that haven't been arbed, will have exploitable inefficiencies(less developed markets/countries).

If your a driver of volatility, you basically end up holding all the cards. Inciters of news events. Look at ES, 2150, 2250, 2350.. consolidation zones (step patterns)..
 
Last edited:
Thanks everyone for your thoughts.
Spectre> my aim was to collect the time value while being as hedged as possible.
My inspiration was an iron condor and trying to improve on it. (Which failed with my original idea)
Will think about your ideas, thank you again.
 
You must log in or register to reply here.
Back
Top