Questrade, My direct access discount broker.

Questrade Democratic Pricing - 1 cent per share, $4.95 min / $9.95 max

Wednesday, June 16, 2010

Erroneous math and a new angle

I made an error in my spreadsheet that ended up using the same account balance for both styles of trade size determination, trade size by value and trade size by contract count. I believe that it stunted the returns in the size by value figures but I changed the starting balance for each quarter so I don't think that it was a huge factor... so I did not run the new numbers again.

Instead I ran them a bit differently today, after correcting the reference error.

Once again starting with $6,700 in each style except that I capped the trade size at 100 contracts. Now the trouble with this is that a $3 option trade is going to cost $30,000 to put on and at that point I would be best trading the stock compared to a $1 option using $10,000 in capital. So I ran a 25 contract cap and a $5,000 value cap.

Contract sizing hit $413,663 in one year and Value capping hit $435,391 for a difference of $21,788 or about 5%.

Back to not making much difference again.

So I thought I would throw a wrench into the works by dropping every 10th trade to not only zero profits but a 100% loss. I doubt that is possible given the methodology we are using but lets compare the results anyway.

Contract sizing starts at 8 contracts per trade based on $25,000 ($6,700 won't stand the losses at that level). By the end of the first quarter the account is down to $23,718. Not only is that a little unexpected that the profits are lower but that the account actually SHRINKS. I don't have to run the rest of the year to see where that is headed. The lowest balance is in the $13,000 area so the account is able to recover somewhat but that sort of a drawdown is not in my books.

Value sizing again starts at $25,000 but ends the first quarter at $34,003 for a gain of $9,000. The account balance dips into the $15,000's but recovers much better due to the equal value of all risk trades. A large option price does not net any larger a loss than a smaller option price.

In the end I think that the value sizing will win out for me as I see the loss mitigation that this method automatically incorporates. Thus a 20% trade on any position is the same dollar profit but also a 20% loss is also the same dollar loss. Having one big $6 option hit zero does not wipe out the gains from six or more $1 trades (12 if they were all 50% gainers). That is worth more than the chance that the $6 option might make 50%.

So in trying to keep trading as simple as possible I checked with my broker to see if they had any easy way to use a dollar amount for trades rather than having to calculate it out then enter the trade quantity... no luck. I know that some US brokers have this but I really don't want to switch. I am going to go back to my trade valuation chart that I set up a while ago when I started this and create a table for quick reference.

It seems that the exits need to be the fastest execution due to the volume of trades closing at the same time driving the option prices down quickly. This is not a problem as I just need to double click on my position then hit SELL. Buying is going to be slowed down somewhat but at least I can pre-set a default "average" trade size and only have to adjust it up or down a few contracts.

This has been prompted by some trades that are going to likely expire worthless this week on me, not that they are big losers but it just got me to thinking about sizing losses similarly as sizing to accommodate losses is at least as important as sizing for gains.

Jeff.

No comments:

Post a Comment