i can already see the responses now....... "GG, stop trying to make everything so complicated!" anyway, i have to bring this up because the concept interests me. it's going to be hard to explain, but here's a theory i have... sometimes i think that each instrument, on every time frame, has it's own average price movement, per time. for example, this can be applied to individual stocks on any time frame. i would argue each stock has it's own price "speed" relative to its timeframe. this can change, but it takes a big event to change it drastically...and usually the change takes place slowly. my point is, things seem to have their own pace. they can go anywhere...up, down, sideways....it doesn't matter. but imo, they don't want to go too far, too fast, compared to how they moved in the past. if it moves too fast, people come in and see it's overbought/oversold for the time and send it back to it's mean price. gaps are a good example. look how many times a stock seems to gap for no reason and gets filled. the move was probably too fast for time and traders come in and fade the move until it is back to what they consider normal. just like there is a different average true range for every stock and it changes depending on the timeframe. i would say there is an average extreme range. price does not want to go x distance away from its mean price before y time. maybe this is what gann meant when he talked about price and TIME. trouble is, i'm no gann expert, so i can't say. even if i'm on to something, i still don't know how to use my idea really....
Gordon, to guess is good, to know is better. why not put this rough idea into tough statistics?. simple start: average halfHourbar next: turn things around and calculate time needed for barwidth. thus: how much time on average do MSFT, AIG, KLAC and C need to create a 1%-, 2%-, 3%-, 4%-Bar? for the programming this would mean: start at each minuteClose and look how many minutes away is the next price that is +-1%,2%,... away from the current price. next: is that timeSpan constant over time or at least the relationship between those stocks? does it change when a fast mving average shows positive or negative slope? is it the same for bars that develop down and for bars developing up? finally you want to be able to create profit, so you have to have tough material, right? peace
here's a chart showing my basic idea..... p.s. albert einstein is well known for spacetime. i will be known for pricetime. just call me gordon einstein.
GG, I see your point. I think your rationale behind it is to look for unusual events, that distort the basic efficiency of the market. I think this is in essence what every trader is looking for, but it is necessary, IMHO, to place knowing on top of guessing. I think it is necessary not to try to reach the goal by the very next step. Your first intention, IMO, was it to distinguish stocks by their "speediness", which most likely will highly correlate with standard volatility. Now you are trying to already place a strategy. I would suggest that you stystemize your thinking. Now as you have that idea with gap versus short term MA. test it out. whenever the market gaps and the opening to MA5 distance exceeds x% (parameter), fade the gap. run it over top30 liquid stocks and you have one strategy - if it works. peace
The trick to being successful is to recognize when a stock "wants" to move faster than average. Good traders know when to jump on the freight train and when not to buy the falling knives. Sounds like you are concentrating on the opposite.
i can always count on gg for getting my mind in gear. very interesting way of looking at price crossing the MA. best, surf
GG - Isn't what you're describing basically just a momentum or rate of change study analytic (whether you apply it to a price chart or a % price change chart)?