Buy at 4pm and sell at 4:30pm strategy

[skc]

I have done some quick numbers on what happens to the SPI (taken from IG Market's ASX200 equivalent which tracks the SPI point for point) between 4pm and 4:30pm. Here are the stats over Sept and Oct:

Total no. of trading days = 45
Total points change for XJO = -1117.6
Total points change btw 4-4:30 = +351.5
Average points change btw 4-4:30 = 7.8 points
Max / min points change btw 4-4:30 = +64 / 55.5 points

I would love to hear from experience traders their thoughts / opinion on these:

1. What can explain the large positive points change between 4 to 4:30? Esp considering how bearish the market has been in Sept and Oct.
2. Can / will you trade on these statistics?

Thanks

 

[chops_a_must]

It's something that I have traded fairly often when on the futures.

Basically, the "theory" is that the end of day trades are often trading counter to the prevailing trend as people close out their positions.

Therefore, most down days of late resulting in people buying back into the futures end of day.

Haven't traded it lately, so couldn't tell you.

 

[barney]

I have done some quick numbers on what happens to the SPI (taken from IG Market's ASX200 equivalent which tracks the SPI point for point) between 4pm and 4:30pm. Here are the stats over Sept and Oct:

Total no. of trading days = 45
Total points change for XJO = -1117.6
Total points change btw 4-4:30 = +351.5
Average points change btw 4-4:30 = 7.8 points
Max / min points change btw 4-4:30 = +64 / 55.5 points

I would love to hear from experience traders their thoughts / opinion on these:

1. What can explain the large positive points change between 4 to 4:30? Esp considering how bearish the market has been in Sept and Oct.
2. Can / will you trade on these statistics?

Thanks

I'm not, and don't profess to be an "experienced" trader (although I've lost a crapload in the past doing dumb things ...lol ...) ........... but, I would say that, what you are describing amounts to "divergance" .......... Possibly the best indicator to a change of trend other than insider knowledge, (from my limited experience) ........... The time frame and range (volatility of price) that the divergance is "working on" (and the depth of your pockets!!) is the key to how you "invest" on it ........... if that makes any sense !!??
PS (Those who have lost a lot of cash and studied the markets may appreciate what I am saying .............. those that have not will simply dismiss my comments .............. Cheers.

 

[skc]

It's something that I have traded fairly often when on the futures. Basically, the "theory" is that the end of day trades are often trading counter to the prevailing trend as people close out their positions. Therefore, most down days of late resulting in people buying back into the futures end of day.

Thanks and like the theory. Although this also happens when the overall market has an up day.

Total trading days = 45, Market up = 14 days, Market down = 31 days

Market up and futures up btw 4 to 4:30 = 13 out of the 14 days (93%)
Market down and future up btw 4 to 4:30 = 19 out of the 31 days (61%)

So in fact this happens at a higher frequency on up days. Granted that the sample size really is quite small.

I would say that, what you are describing amounts to "divergance" .......... Possibly the best indicator to a change of trend other than insider knowledge, (from my limited experience) ...........

I thought divergence require two things to diverge - like price action vs an indicator. What is the divergence you are referring to here?

And the SPI rising btw 4-4:30 definitely hasn't seen the market rise the following day (as the data above suggests). Do you mean a short term trend reversal indicator?

The time frame and range (volatility of price) that the divergance is "working on" (and the depth of your pockets!!) is the key to how you "invest" on it ........... if that makes any sense !!?? PS (Those who have lost a lot of cash and studied the markets may appreciate what I am saying .............. those that have not will simply dismiss my comments ..............

I definitely don't have lot of cash so have no idea what you are saying. Would be great if you can share more.

cheers

 

[skyQuake]

You might wanna try 4:10pm to 4:30pm. Thats when cash closes proper.
Recently there would have been a whole heap of stocks being dumped by aussie and international funds from rebalancing, redemptions and whatnot. This has a negative effect on the Cash, and futures are dragged down by the Arb bots. Post cash close people look for bargains? Or without the selling pressure the market simply rises.
Interesting indea, but I think you need a far bigger sample size to test it out properly. Who knows when the reverse will happen?

Good luck

 

[barney]

Thanks and like the theory. Although this also happens when the overall market has an up day.

Total trading days = 45, Market up = 14 days, Market down = 31 days

Market up and futures up btw 4 to 4:30 = 13 out of the 14 days (93%)
Market down and future up btw 4 to 4:30 = 19 out of the 31 days (61%)

So in fact this happens at a higher frequency on up days. Granted that the sample size really is quite small.



I thought divergence require two things to diverge - like price action vs an indicator. What is the divergence you are referring to here?

And the SPI rising btw 4-4:30 definitely hasn't seen the market rise the following day (as the data above suggests). Do you mean a short term trend reversal indicator?



I definitely don't have lot of cash so have no idea what you are saying. Would be great if you can share more.

cheers

Hi SKC, Firstly I must state that I wasn't trying to be clever or flippant with my comment ............... and my statement that I am not an expert trader is very accurate (although I have had a lot of experience trading .... most of it negative ... lol)

Short term is definitely what I was referring to with the "divergance" ........... If the market is tanking (as it has been) and the SPI has been rising significantly EOD, then the question should be asked ... who is "buying the Index" .............. Logically the EOD may often be opposite the "trend" as short term holders close out their positions as Chops pointed out etc.,............
............. but if the "trend" is so obvious, why does the smart money simply not hold their positions and "cash in" overnight?? .................. Because they are unsure whether the market will rise X% or fall Y% !! Volatility to me indicates lack of definition .......... The fact that the market can increase so substantially in the face of such negative sentiment means .... be cautious ........ and don't assume that the trend is continuous ............. It certainly doesn't mean that it can suddenly turn on a sixpence and skyrocket (although it might just do that) .................. ie. If the smart money is being cautious, then the downtrend is fragile imo. ........... As the downtrend becomes more fragile, their is more chance that the bottom will show itself ....... Hence my reference to "time" and the "depth of ones pockets" being relative to ones "investment" plan ...............

If you have limited cash available in the current market (join the club ) then only trade very short term, or alternatively, only trade very long term (ie be prepared to wait) cause things may or may not get a lot worse .......... PS ... Just my observations from watching the market over a long period, and I may be totally off the ball .................... The point is you have picked up on an anomaly of the markets that you may be able to "take advantage" of ............. This may or may not be a short term anomaly .......... but as long as you/we are aware that things in the market are not static, we may be able to derive some advantage from them. Hope that makes some sense ......... I've pretty much confused myself ... lol ..... Cheers.

 

[skc]

You might wanna try 4:10pm to 4:30pm. Thats when cash closes proper.

Yeah. I wanted to do 4:10 to 4:30 as well. But my 30min data goes back further than my 10min data. Also, I wasn't sure if the arb bots could work in the closing auction, given that they can't really place a "market order" per se.

Interesting indea, but I think you need a far bigger sample size to test it out properly. Who knows when the reverse will happen?

Will post some Nov data at end of month.

Hi SKC, Firstly I must state that I wasn't trying to be clever or flippant with my comment ............... and my statement that I am not an expert trader is very accurate (although I have had a lot of experience trading .... most of it negative ... lol)

Hey Barney, didn't think you were trying to be clever at all . Appreciate your input and just interested to understand more / better.

The point is you have picked up on an anomaly of the markets that you may be able to "take advantage" of ............. This may or may not be a short term anomaly .......... but as long as you/we are aware that things in the market are not static, we may be able to derive some advantage from them.

To trade profitably from this one will need to consider range and stop loss. There are times when it drops 20-25 points before going back up to finish 1 point above the line - not the best R/R ratio and I would porbably have closed the position at a loss, even though the stats said you made a profit.

And don't we all have "limited" cash. I probably wouldn't be trading if I had "unlimited" cash

 

[Broadway]

Asia still has some influence over the spi at this time.

The kospi and nikkei are still open at that time (at the moment) and these can run parallel. The spi isn't an isolated independent index. If you know where the Asia region is trying to go you can predict the spi.

And in winter try putting a chart of the dax next to the spi at 4pm.

Good luck with the strategy, hope it works for you.

 

[MRC & Co]

If you know where the Asia region is trying to go you can predict the spi.

lol and sometimes this will get you RAPED!

 

[superkrusty]

SKC , would you mind letting me know your method or process of checking this , I have a few similar ideas I want to test but have no idea on how to do it. Was thinking of using excel ?? Am I on the right track?

 

[MRC & Co]

lol and sometimes this will get you RAPED!

Today would be a fukcin good example!

 

[professor_frink]

SKC , would you mind letting me know your method or process of checking this , I have a few similar ideas I want to test but have no idea on how to do it. Was thinking of using excel ?? Am I on the right track?

yep. Throw the relevant data into excel, then you just need to know how to code the idea into it. You don't need to be much better than a beginner to start testing out ideas like this.

You can also do it with most decent charting packages(amibroker, metastock,etc), but you'd be surprised at how much you can actually do with a simple spreadsheet once you know how to use it.

 

[awg]

would not have worked today! on either of the below

are you using the "Australia 200 forward"

or "aus cash 200"

interesting idea though, where are u getting your historical data?

thanks

 

[skc]

SKC , would you mind letting me know your method or process of checking this , I have a few similar ideas I want to test but have no idea on how to do it. Was thinking of using excel ?? Am I on the right track?

yep. Throw the relevant data into excel, then you just need to know how to code the idea into it. You don't need to be much better than a beginner to start testing out ideas like this.

You can also do it with most decent charting packages(amibroker, metastock,etc), but you'd be surprised at how much you can actually do with a simple spreadsheet once you know how to use it.

Spot on Professor. You will find endless ways to analyse the data if you get familiar with "Pivot Table" function in excel. That is also quite a beginner-level tool.

would not have worked today! on either of the below

are you using the "Australia 200 forward" or "aus cash 200"

interesting idea though, where are u getting your historical data?

I looked at Aus cash 200. Just read off the chart for historical data - there might be smarter ways of doing this, however.

The limited observation so far merely suggests that there is a higher odd of rising than falling during these hours. Make of it what you will.

Today's number are: 4pm = 3532.5; 4:10pm = 3529.5; 4:30pm = 3535. So there is a small gain, but you would have had to endure a low of 3502 at 4:15.

 

[awg]

Spot on Professor. You will find endless ways to analyse the data if you get familiar with "Pivot Table" function in excel. That is also quite a beginner-level tool.



I looked at Aus cash 200. Just read off the chart for historical data - there might be smarter ways of doing this, however.

The limited observation so far merely suggests that there is a higher odd of rising than falling during these hours. Make of it what you will.

Today's number are: 4pm = 3532.5; 4:10pm = 3529.5; 4:30pm = 3535. So there is a small gain, but you would have had to endure a low of 3502 at 4:15.

Just for my clarification, as I am only CFD learner

I thought the aus cash 200 was a mirror of the ASX200, and the aus 200 forward was a mirror of the SPI?

dont want to seem to be a smartars but with a 2 point spread each way, would those figures not suggest a tiny loss?

ps sure makes a difference paper trading on a demo account, I absolutely cleaned up today, certainly would not have the guts to trade like that for real at this point in time.

keep up the great work

 

[James Austin]

SKC

where is entry? 4.01pm!
where is exit? 4.30pm!
where is SL placed? none! . . . eek!

i dont wish to discourage you, but these things need to be sorted.

i am pretty confident that a large sample will, on the SPI b/w 4 + 4.30pm, reveal a random pattern, but there might be some clumping of up days and dn days. the problem is, no one knows when the clump will start and when it will finish; if you can work that out just place $1000pnts and make a killing!

returning to entry, exit and stops,

even if you come to the conslusion that 60% of the time the SPI is up b/w 4 and 4.30pm, you still need to know where to place entry and exit, cos the SPI doesnt travel in an upward straight line as of 4pm, . . . how will you work this out?

and if you decide well i've got a 60% chance of profiting here so i'll just enter at 4.01 and exit at 4.29, . . . . where is your stop, you might need a 50 or 60pnt stop to allow for the volatility at this time of day.

in the end, all of this is easily and cheaply tested, $1 pnts at CMC or City Index.

i await the results

 

[skc]

I thought the aus cash 200 was a mirror of the ASX200, and the aus 200 forward was a mirror of the SPI?

I think the cash and the forward are tick-for-tick anyway. So since we are looking at the differences it should produce the same outcome regardless of the instrument.

dont want to seem to be a smartars but with a 2 point spread each way, would those figures not suggest a tiny loss?

It's actually 2 points spread total. The numbers quoted are mid-point of the spreads. So to buy you pay 1 point more, to sell you get 1 point less. So for today, there is 0.5 point increase.

where is SL placed? none! . . . eek! i dont wish to discourage you, but these things need to be sorted.

Totally agree James. See post #7 above. I am also well aware of the small sample size. I am not trading this strategy (strategy's probably too big a word, more like observation) - it is being very cheaply tested on paper

The purpose of this thread is to ask if anyone has any reason explaining the observed. I don't believe in the Asian market influence theory just yet - it doesn't make sense that Asian markets should rise btw 4-4:30. I will probably do some correlation testing if there are theories that stack up... Do share your thinking.

cheers

 

[awg]

one observation, may mean nothing

but yr sample roughly corresponds with short selling ban

will be interesting to see what happens when short selling is allowed, as yr time period (4 to 4.30) is also right when short sellers would be very active

 

[skc]

Just a quick update on the Nov numbers as promised - for anyone still interested.

Total no. of trading days = 20
Total points change for XJO = -479
Total points change btw 4-4:30pm = 105.5
No. of days XJO went up btw 4-4:30 = 14
Average points gain btw 4-4:30 = 7.5

Total up ratio for last 3 months = 46 / 65 days (~70%)

Again, feel free to discuss any plasible explanations.

 

[awg]

keep up the good work

if you are keeping data, what would be interesting in addition, would be downside volatility.

ie R/R for the 7.5 up points

when i get time, i will have a look myself

can u remember what the post title was for that dude who was trying to scalp 1 point with a 20 point stop?