$250,000 Portfolio

First of all, to get a sufficient Return on Capital (ROC), we should use leverage.  A normal margin account gives 2:1, so let's start there.

250,000 times 2 means we are controlling $500,000 worth of ‘stuff'

Let's pick some products that may not be correlated to give us some diversification.  Let's pick our favorite direction for each i.e. long or short.  It turns out that it does not matter what we choose in the long run as all of these decision fall into the 50/50 category.

We call this our "Core Portfolio"

If we are long some ‘stuff', then we sell Call Options against them to reduce our Cost Basis and improve our odds of winning.

If we are short some 'stuff', then we sell Put options against them to reduce our Cost Basis and improve our odds of winning.

It would look something like this:

The following are considerations for what we sell against our 'Stuff':

     1. Selling one Std Dev OTM options gives good balance between 'enough' premium and amount of management needed

          a. should be right about 68.5% of the time - in reality more like 75-80% of time

          b. 1 SD yields about 1%/trade a month or so out  on a typical underlying with typical IV - since each trade has a high probablity of success, it works out to just about 1%/month RoC (unleveraged)

          c. ATM yields about 4%/trade , but you are wrong about half the time which leads to more management - half wrong means closer to 2%/month average and more management costs means actually something less than 2%/month RoC (unleveraged)

               - testing shows a slightly better Return, but not really worth the effort for most people

     2. 1 SD = 1%/Mo = 12%/year target as a reasonable Rate of Return to shoot for - No Leverage

     3. A typical Margin account has 2:1 leverage which would yield a target of 2%/month or 24% annulized Return on the Core or a 'Cost Basis Reduction of almost %25 - In 4 years, the 'Stuff' has been paid for by the sale of premium

In our example, 1% of $500,000 would be about $5000 in premium/month which is 2% of $250,000 

     4. The Core will go up or down depending on if your 'guess' was right or not; but the benefit of this method is the Cost Basis Reduction gained by the selling of premium

So, in the above portfolio, we used about $160,000 in buying power to control the $500,000 worth of stuff - since we started with $250,000 in this case, this leaves about $90,000 of unused capital.  Since we are leveraged, and also to save some for future opportunities or problems, we don't want to use all of this 'extra' capital.

Let's assume about one third is used, that is about $30,000.  We will use this to make other trades 'around' our Core Portfolio.  Here are some considerations:

     1. In order for the probabilities to have enough space to normalize, we need many occurrences - maybe 300 or more annually

     2. Each occurrence should be small so as to reduce the risk of a single trade going against us having too much of an impact on our portfolio (Security Risk) 

     3. We should have underlyings that are not correlated so that each is a true 'occurence' and not just a 'bigger' occurence 

     4. We should be long and short different things to mitigate a bit of risk (Market Risk)

     5. Maybe a 'Delta Neutral' or 'Lean Short" or 'Lean Long' type of mentality 

The Trades could be split between Defined Risk and Undefined risk as a good balance 

     1. Undefined Risk have a higher winning percentage and yield, but carry more risk

     2. Defined Risk are not as efficent, but carry a known Max loss amount

Here are examples of each: