Are indices random intra-day?

[tomorton]

I have been watching the intra-day movement of the FTSE100 virtually every day for the last 7 years. I have also been watching its EOD prices, OHLC, and longer-term trends, levels and patterns. Not an expert, but not an amateur.

It remains very hard for me to predict with any reliability the O, H, L or C tomorrow, and certainly not the intra-day s/r levels lying between H and L. Sometimes they are flagged up by past charts, sometimes they are not, and never sufficiently consistently to make me think it's anything other than chance.

I had sudden inspiration earlier though - I had been viewing the intra-day market behaviour as just a smaller-scale picture of the month or year's perfofmance, so that one day builds on the one before, in linear fashion. If that were true, by knowing more and more about the last 249 trading days in 12 months, it would be increasingly easier to predict salient price behaviour in Day 250.

Actually, maybe the truer picture is that the year's 250 sessions are not a linear series, leading to a price on 31/12, but they are actually parallel and entirely random. The new picture is not of a road, and a driver who predicts the next bend by using the rear-view mirror and his experence of the road's past typical curves and bends and straights - but it is more like a herd of 250 migrating buffalo, who all have to cross a trackless plain, side by side, all starting at the same time. (This is an odd image for technical analysts, as time is re-arranged from linear to parallel, when our everyday experience tells us that tomorrow is a continuation of today.)

Naturally, you'd say it's impossible to say where each buffalo will go with any accuracy as soon as they have gone more than 6 feet from the start line. They may drift, veer right, they may sit down for half the day, chewing the cud, they may stampede straight ahead, then slow down, turn left and drop down for a sleep. There's no way of knowing - and observing 249 of the buffalo will do nothing to predict what the 250th will do.

Now that's how I view the FTSE.

 

[scose-no-doubt]

http://www.trade2win.com/boards/general-trading-chat/118256-another-scose-type-question-thread.html

 

[Pi3141]

I think for intraday traders you dont need to have a longer view on what the markets may or may not do. Therefore, imo, it's perfectly fine to approach the markets everyday without any bias and as if they are random.

Tomorton, may I ask why you've decided to trade the FTSE for so long? What is it about the FTSE that attracts you to trade it? I'm just curious because I always assumed UK traders would start off with the FTSE at the beginning and then move on to other markets later on.

 

[Rhody Trader]

It remains very hard for me to predict with any reliability the O, H, L or C tomorrow, and certainly not the intra-day s/r levels lying between H and L. Sometimes they are flagged up by past charts, sometimes they are not, and never sufficiently consistently to make me think it's anything other than chance.

Random and unpredictable are not necessarily the same thing.

I had sudden inspiration earlier though - I had been viewing the intra-day market behaviour as just a smaller-scale picture of the month or year's perfofmance, so that one day builds on the one before, in linear fashion.

I'd say fractal - as in short-term patterns are parts of larger patterns - but not linear. Markets are most definitely not linear.

Naturally, you'd say it's impossible to say where each buffalo will go with any accuracy as soon as they have gone more than 6 feet from the start line.

With enough information you could predict, but it would take masses of it and loads of computing power. Again, hard/impossible to predict is not the same as random. A buffalo won't randomly stop when it's running. There's a reason for it doing so. We just may not know what that reason is.

 

[rawrschach]

Everything is random. The question is how and how much.

 

[Tubbs]

I've actually come back from the NQ to the FTSE just because of the movement. My system relies on the market moving a certain way. for example it is good for the market to break, and it it retraces to not do so pass levels of large volume. The FTSE generally obeys this rule more than other indices like Nasdaq or S&P.

 

[Splitlink]

Random and unpredictable are not necessarily the same thing.



I'd say fractal - as in short-term patterns are parts of larger patterns - but not linear. Markets are most definitely not linear.



With enough information you could predict, but it would take masses of it and loads of computing power. Again, hard/impossible to predict is not the same as random. A buffalo won't randomly stop when it's running. There's a reason for it doing so. We just may not know what that reason is.

Something like MP? I spent some time on that but came to the conclusion that it did not resolve what I wanted to know, which is market direction. Buying cheap and selling high is, always, what a trader should be trying to do. If it is not possible to do that but only buy at the extremes in the hope that it will go the other way for a better price, then that is controlled random trading with a low risk to reward factor because it should move to the fair value area, at least.

 

[Nexas]

I don't believe the markets are random as a whole... I do believe that you cannot predict the very next movement or very next trade... but then do you need to in order to make money? Of course not.

So I would say that anyone buffalo's movement is unpredictable and random but ultimately by watching the movements of many you should be able to get an idea as to where the pack/heard are going...

So as with trading, any one trade is random and completely unpredictable however over a large enough sample size you will see patterns which repeat themselves which is enough to make profit

I hope that makes sense

 

[Splitlink]

I have been watching the intra-day movement of the FTSE100 virtually every day for the last 7 years. I have also been watching its EOD prices, OHLC, and longer-term trends, levels and patterns. Not an expert, but not an amateur.

It remains very hard for me to predict with any reliability the O, H, L or C tomorrow, and certainly not the intra-day s/r levels lying between H and L. Sometimes they are flagged up by past charts, sometimes they are not, and never sufficiently consistently to make me think it's anything other than chance.

I had sudden inspiration earlier though - I had been viewing the intra-day market behaviour as just a smaller-scale picture of the month or year's perfofmance, so that one day builds on the one before, in linear fashion. If that were true, by knowing more and more about the last 249 trading days in 12 months, it would be increasingly easier to predict salient price behaviour in Day 250.

Actually, maybe the truer picture is that the year's 250 sessions are not a linear series, leading to a price on 31/12, but they are actually parallel and entirely random. The new picture is not of a road, and a driver who predicts the next bend by using the rear-view mirror and his experence of the road's past typical curves and bends and straights - but it is more like a herd of 250 migrating buffalo, who all have to cross a trackless plain, side by side, all starting at the same time. (This is an odd image for technical analysts, as time is re-arranged from linear to parallel, when our everyday experience tells us that tomorrow is a continuation of today.)

Naturally, you'd say it's impossible to say where each buffalo will go with any accuracy as soon as they have gone more than 6 feet from the start line. They may drift, veer right, they may sit down for half the day, chewing the cud, they may stampede straight ahead, then slow down, turn left and drop down for a sleep. There's no way of knowing - and observing 249 of the buffalo will do nothing to predict what the 250th will do.

Now that's how I view the FTSE.

Sorry, Tom, but I do not believe that the markets can be predicted reliably but that they move for different reasons, all the time. All I try to do is get the entry right but, once into the trade, it can,and does, move in a random manner IMO, and we can never be certain why it has moved in the direction that it has, even though we think we do, afterwards.

I know, from previous threads, that you are a student of TA and have not tried to disuade you from that, every man to his own poison, but, since you have brought the matter up........ I don't think that there is anything new that hasn't, already been discovered by now.

Look at a daily bar. It may be trending up or down but, between the open and close, no one can predict how many ups and downs there will be before EOD.

 

[PurplePerson]

Sorry, Tom, but I do not believe that the markets can be predicted reliably but that they move for different reasons, all the time. All I try to do is get the entry right but, once into the trade, it can,and does, move in a random manner IMO, and we can never be certain why it has moved in the direction that it has, even though we think we do, afterwards.

I know, from previous threads, that you are a student of TA and have not tried to disuade you from that, every man to his own poison, but, since you have brought the matter up........ I don't think that there is anything new that hasn't, already been discovered by now.

Look at a daily bar. It may be trending up or down but, between the open and close, no one can predict how many ups and downs there will be before EOD.

I agree. It is possible to spend thousands of hours and thousands of pounds devising a trading system that attempts to refine and refine and refine predictability. As if randomness is like radio interference, which the expert radio engineer can more or less eradicate to reveal the desired signal. I consider it more honest to acknowledge that randomnesses are an intrinsic element in the movement of market prices. That it's better to acknowledge that rather than deny it and be thrown off course by it when it kicks in. For me, a system which allows for and copes with random movement is a lot safer than a system which attempts to play down its existence.