Sorry to be late to this thread, but check out this link: http://www.smartquant.com/introduction/openquant_strategy.pdf It is a basic beginner manual on how to build strategies using the OpenQuant system. The relevant part is found on pages 7-12: Summaries of some popular kinds of trading systems. Through p.16 it gives IMO good basic general advice on designing systems, then follows with examples of the code for common systems. I think it would be a good place for anyone to begin and "see the kinds of structures of these strategies and use them as a foundation."
occam's razor of course. thomas edited my oddball article in the dec 2000 active trader magazine. i am very greatful to thomas and have great respect for him. also he is the person who named oddball. we were discussing what to name the article and he said this is sure an oddball system and so i said why not just name it that and so we did. chicago is a great place and many smart guys gravitate there. but no matter how smart everyone wants what i call a pocket system. a system that you can in your head or on a piece of paper in your pocket keep. that is why pivot points and all sorts of other support resistance calc's are so popular with guys in chicago. i dont particularly like those types of calc's but i have respect that prices do gravitate to those zones, especially fib numbers. i don't use them at all but i have to say i think i could use them to trade and they are simple enough. beyond that i like market breath as anyone knows. i also like tracking strength differences between indexes. a guy could track differences between intraday highs and lows, making a dynamic center pivot point. there are just so many simple ideas that with good money management and an unwavering ability to flip a trade doubling up when it's gone wrong. i really don't know where to begin as i set here thinking. i will tell you why i think most people never get it. they want to make money so bad that they will cheat the discovery process. i mean using trailing stops in ts or some bogus daily data system where the average trades in looser and winners is 0 etc. there are so many fallacies in ts and other programs that just that alone is a mine field. i think the other problem is most people lack the deep history of data to properly test their ideas on. myself and one other developer require 20,000 trades to validate a model. then you have bad data, slow feeds, you can begin to understand why simple methods are better. i do think that myself. i had a method once where i could simply get the number of advancing issues on the nyse and from that one number i could trade the sp. think not even using the sp data i could trade the sp. powerful stuff. as far a having a mantra "the minority make a fortune doing the opposite of the majority". i am a counter trend trader pure and simple. meaning to me a trend is an extended move in one direction. i always fade the move initially and only go with the trend on a stop and reverse after the counter trend trade has failed. mark brown