I'm not referring to systematic as full-on 100% automated algo trading, but clearly defining a ruleset that has been backtested, and you will not deviate from it. At least not in the moment... Discretionary as in - Here is what I for look for (maybe a general rule set) but from there is subject to interpretation from what I see going on and the current market environment.
i have a defined process that i use for trading starting with baseline expectations, context, setup, entry, trade management and post trade analysis... some parts are rigid and other parts use fuzzy logic. i don't backtest anymore and test all changes by forward testing using sym first then in real time.
THe Jackson 5 says systematic trading is the way to go. Ah baby Move it baby Automatic, systematic Full of color, self-contained Tune that channel to your box Captivating, stimulating She's such a sexy lady Filled with space age design She's moving, she's grooving Dancing untill the music stops, now Yeah
This is a misleading premise. Many people think they have a "system" but what they have are a bunch of ill-defined rules, that are not actually rules at all, but more like guidelines. (is that a quote from some comedy movie?) "Clearly defined ruleset" are scriptable-able. If they are not, then by definition, they are not clearly defined. And-Or worse, they are incomplete, meaning there are cases where there are no existing rules but should be. One can't actually backtest, in a meaningful way, these ill defined "system". I would council the OP to actually write up some rules in a scripting language: When and how to enter. When and how to exit. Probably it will be an eye opener that these "Clearly Defined Rules", are so badly defined that they are incomplete (leave out too many situations), or so basic, they have no reasonable profit. Another thing is these rules are usually created by cherry picking a few setups that had positive outcomes. That leaves out when they: go straight to negative, when they go negative and then recover, or are positive and then go negative, or go nowhere and the market closes. Do your "rules" cover all those situations? If not there is work to be done. OTH, just go discretionary, and pretend that you are experienced enough, technically, logistically, operationally and emotionally to master it. Then, try to sustain it for a couple of years. Not recommended for full time serious traders except as a "tour of duty" with support.
i have an automated trade management system so 1-2 days for any changes. for setup changes, i watch and manual test for more than 6 months before changing anything... no sim trading for setups, just jump right in assuming that after 6 months the change will improve profits.
%% Good discretion /about your heading higher comment days ago. BUT no such thing as a good system without good discretion. I like much larger time frames than a moment; even though many of those make a good trend...................................... MORE wise rules the better; dont have any rules that are unwise NOT a prediction
Most successful discretionary traders have a system that follow religiously. When the system is matured it can be automated. I don't really know how someone could trade without a system or some kind of a plan.
Trade plan, system, logic, whatever you called it, everyone has one. It is not randomized entry or exit.
It can probably be partially automated. I have a discretionary part and a systematic part. The discretionary part is there because that part cannot be automated. There are millions of different situations possible and for optimal return it is impossible to automate these millions of situations. I tried many years to write down all the possible situations/rules, but never succeeded. In each trade that I do there is a discretionary part, so trading full automated is impossible. The discretionary part is an essential part of each trade. In an automated system, a certain tolerance must be built in to prevent too much potential return from being lost. If the rules are too strict, many signals that just fall short will be lost. The result will be that the return will be much lower. The discretionary part is more flexible and improves the return. Just like defining all the rules for driving a car is impossible.