Here we go again!
It looks like I have generated some interest in the methods I use, so I will begin giving some specific info on the things I have learned over the years.
Before we go any further, let me repeat what I said previously. Learn math. It is the LANGUAGE of investing, and without it you are crippled. Learn to calculate in your head, with a reasonable amount of accuracy. If you trade like I do, you will very often not have time to pick up a calculator and key in numbers. Practice.
Learn to handle Excel well. I don't mean all of the esoteric stuff, but I use the usual simple calcs, powers and roots, and some charting.
For testing your ideas, you will need to process LARGE, LARGE amounts of data. Don't invest in SQUAT unless you are certain the odds are on YOUR SIDE. You will only know this if you have PROVEN you theses by submitting them to rigorous testing. If you don't test, you are flying blind. and run about a 50/50 chance of losing.
As part of your brushup in math, take a hard look at SIMPLE PROBABILITY and MATHEMATICAL EXPECTATION. Every good horse player knows this stuff, and you must as well.
----------
I got some emails that I should answer, so here goes.
rtiger29 -- re: Measurement of success
As I said bedore, I don't use dollars gained as a measure. It's fun, but not really productive. Here is the formula I use:
(SellPrice/(BuyPrice + BuyCommission + SellCommission)^(1/NbrDays)
This is the way Excel wants to see it entered. If you are using another language, you may have to modify.
Now, what does this give us? Note that the commissions are added to the per share cost at the beginning. That way, the output represents PURE PROFIT. Here, we are dividing the SellPrice by the total costs to get the ratio.
But remember, we need to include TIME as a factor.
Years ago I arbitrarily set a goal of 2% for each trade per day. I knew it wasn't a realistic target, its far too high. But, it's something to strive for.
When we compound that, using 250 market days in a year, we get
1.02^250
(we divided 2% by 100 giving .02, added 1, and raised to the 250th power)
The result is 141.268. What that means is that if I had bet $1 and won 1.02 times that much, then bet $1.02, etc, etc, until I had done it 250 times successfully, I would have $141.268.
As you can see, this is wildly optimistic. However, it is ONLY a target. Actually my current rate is a little under 1.01 for the last 20 trades. It keeps me happy for now.
So, to include TIME as a factor in the formula above, we take a ROOT, that is 1/NbrDays, where NbrDays is the number of days you held the position. This appears in the eq as ^(1/NbrDays).
----------
tyrant -- re: 70% Wins
In most of the books I have read about trading, a 70% win rate is considered good. Actually there are TWO FACTORS involved, the WIN RATE and the GAIN RATIO.
Example: I bet $1. I have a 20% chance of winning. But historically I know that my average gain ratio is 10 to 1, i.e. win $10 for every $1 I bet. Look at the numbers: I have a 1 in 5 chance of winning 10 times what I bet. Good bet.
What I am saying is that you must consider BOTH of these factors in order to bet logically. If you only win 50% ot the time, then you must seek out positions where you are assured of making at least $2 for each $1 you bet. Anything above that will be profit in the long run.
Now, obviously, searching for opportunities like that might be fruitless. So, you must concentrate on three things:
(1) Raise your WIN RATE.
(2) Raise your GAIN RATIO.
(3) Improve your skills at searching for information which benefits your investment program.
Example: I spent hours today looking for info on "Rare Earth" opportunities.
Now, I know that Molycorp (MCP) is in the process of reactivating mining interests in lower Calif. Why would I waste time on Sunday doing crazy stuff like that? Because I know that that mining area was closed down some time ago, and the US substantially withdrew from the global market because of environmental concerns. I also know that China has rich deposits of rare earth minerals, and they appear to be on a global commodity binge. So, I will watch MCP from time to time, for any sign that an opportunity is there.
You make your luck.
----------
dwpeters: -- re: Swing Trading
Take one of your longer trades, and convert it into a compounded rate per day as I have done above. Lets say you bought at $10 (ignore comms) and sold 25 days later for $12.50. When you enter that data into the eq, you get a daily compounded rate of 1.009 approx. That is roughly my average right now, but it takes me FAR LESS TIME to accomplish the same end. This means that I have funds free, not tied up, that I can use for other opportunities. If you look at the math, you are ALWAYS better off to trade SHORT TERM than long term. Try out some calcs.
So Long for now, guys.
Good Luck!
