Market Statistics - Finding a statistical Edge in your market

[CanOz]

People might misunderstand the purpose of knowing your statistics....it is as much for keeping you out of a trade as it is for getting you in one. If the market has been one time framing it and it's only put in half its daily average range, one would have zero chance at a successful fade even if it's at a solid level....as an example

 

[cynic]

....
Next step is to find out if changing the start day of the week has any significance.
...

Hopefully look at some other markets as well

Ewe write two bead out full!

Buy inn tent sieve eggs acute shun off algae rhythmic comb pewter proses, eye fine alley no watt two eggs pecked.

Weak bee kin old tars fine all rear salt!



Pea erse:7500 daze SP500 dart tar eggs salmon neigh shun.

 

[ThingyMajiggy]

Ewe write two bead out full!

Buy inn tent sieve eggs acute shun off algae rhythmic comb pewter proses, eye fine alley no watt two eggs pecked.

Weak bee kin old tars fine all rear salt!



Pea erse:7500 daze SP500 dart tar eggs salmon neigh shun.

 

[CanOz]

Not sure what you were drink'in last night Cynic but let me know where ya bought it...

Looking at the HHI now that i have some intraday data...

I've always suspected that the first 15 minutes prior to the cash open might be significant.

We can test a range breakout here with a half range target and see what % of the time the system would achive a half range target after breaking out of the pre-cash range.

A 69.64% win rate tells us that a break of the range will result in continuation to a target of at least half the range.

58% of the time the market will achieve 100% of this pre-cash opening range upon breakout.

51% of the time the market will achieve a target of 2 x pre-cash period.

Now while i have no intention on trading this as a system it is almost trade-able as a system on its own. it has a very low draw-down

 

[CanOz]

So how does this help the discretionary trader?

1. Don't try and fade the opening range

2. Look for the potential of the move, what obstacles are in front of a move out and past the range? Set reasonable targets? Some of things i'm thinking of in order to put this in the book as a play.

3. Because i used a filter as well, the trade depends on an expansion of volatility that is setup before hand. I need to be patient for this trade to work the best.

 

[CanOz]

Also, consider how a breakout of the range, either long or short fits in with the overall context, technical, fundamental/news.

Does the market open inside or outside of its current bracket or balance area?

Imbalanced opens then to be more volatile.

 

[ThingyMajiggy]

Not sure what you were drink'in last night Cynic but let me know where ya bought it...

Looking at the HHI now that i have some intraday data...

I've always suspected that the first 15 minutes prior to the cash open might be significant.

We can test a range breakout here with a half range target and see what % of the time the system would achive a half range target after breaking out of the pre-cash range.

A 69.64% win rate tells us that a break of the range will result in continuation to a target of at least half the range.

58% of the time the market will achieve 100% of this pre-cash opening range upon breakout.

51% of the time the market will achieve a target of 2 x pre-cash period.

Now while i have no intention on trading this as a system it is almost trade-able as a system on its own. it has a very low draw-down

Good stuff Can going to be a gem of a thread if it keeps going.

You done anything similar on the Eurex products from their hour pre-cash session? How much intra-day historic data for FESX have you got?

 

[CanOz]

Good stuff Can going to be a gem of a thread if it keeps going.

You done anything similar on the Eurex products from their hour pre-cash session? How much intra-day historic data for FESX have you got?

I had yes, i think there is some data here somewhere on those tests...i'll have a search.

 

[CanOz]

I had yes, i think there is some data here somewhere on those tests...i'll have a search.

I can't find that other thread at the moment but that strategy that i tested has a filter that only would trade the opening range (pre-cash) if there was a gap up or down, so the market needed to be imbalanced. It worked quite well from memory but i don't have that code anymore. My current code uses a filter as well, but its not a gap. It doesn't test very well on the 60m pre-cash open at all, or any other duration for that matter. With or without the filter. If anything you'd be better off fading the range...not sure why that is.

 

[CanOz]

Hers the other thread where I was testing the gap orb strategy https://www.aussiestockforums.com/forums/showthread.php?t=7046&page=20&highlight=Scratch

 

[fiftyeight]

Ahhhhhh, why in a modern 24 hour world, with 24 hour news and markets with ETH, why are the stats showing what they are showing haha

Donated by a very helpful fellow ASF member

CL


NKY


FDAX

 

[CanOz]

Ahhhhhh, why in a modern 24 hour world, with 24 hour news and markets with ETH, why are the stats showing what they are showing haha

Donated by a very helpful fellow ASF member

Awesome, this is great.

So now we know the probability for a high or low for the week and the time of day, or did we do those markets?

 

[ThingyMajiggy]

Interesting, so basically more often than not, every week plays out like the shape of a tick, Monday more likely to be the low of the week, pushing higher through the week to a more than likely high on Friday? Is this because we've been in a bullish market? Wonder if it's the opposite for bearish periods.

 

[fiftyeight]

This data goes back to 2009. It should not be too hard to retrospectively go back and run it for only bear markets during this period or for ranging periods.

 

[fiftyeight]

Awesome, this is great.

So now we know the probability for a high or low for the week and the time of day, or did we do those markets?

My additional charts are from daily driver. The only hourly data I have is from the SPI that I got from AOTB

 

[fiftyeight]

Really quickly while I eat.

Zoomed out the NK is ranging from 2009 - 2013, however the shape of the most probable high remains very consistent.

 

[CanOz]

Today I'm going to do some research on the question "Are the Thursday and Friday at the first of the month rotational days before the announcements (ECB/NFP). If it is true then we should be able to know that we can safely apply 'balance' trading rules, i.e. fade the range extremes.

To do this i'll pull up a couple of longer term FESX charts and simply look for rotational price action prior to the announcement times. Rotational being highs and lows but no real trend, price basically unchanged into the announcements. I'll start a spread sheet to help quantify the results. Now this will not be enough info to make a decision on, but perhaps enough to go further with the research.

I'm also going to answer another question as well "Did the market open in balance or out of balance?"

 

[CanOz]

So, what is the price action like on the first two days of the month prior to the big announcements, ECB and NFP?


Just going over the period from June 2015 to yesterday, you can see that in almost all cases, if the market opens in balance, then it will rotate around until the announcement, on both days. To explain 'in balance' means opening in the prior days range basically. If you have volume profile then it should be the prior days value area. There are times when the market can open outside the prior days balance area but still behave as a balanced market, usually because it opens in a prior balance area, which all makes sense when you think of the market in terms of accepting value.

Also, there was one day that ended up being a trend day, but there was some jawboning on that day regarding the ECB's QE program.

So this tells us what most veteran traders would likely already know anyway:

Don't be looking for big trendy moves, fade the value area highs and lows for trades back to value and the other high or low without the fear of getting run over. Its good context for the day.

I'll probably go on and do some research around these two days and a couple of others, for example...Monday's, if they open in balance do they trade rotationally until the US opens? If they Gap open,do they always trend?

I likely won't post all the results but its just to give you an idea of ways to look for plays and context for those particular days.

Hopefully, others can post some research of their own...

 

[fiftyeight]

So, what is the price action like on the first two days of the month prior to the big announcements, ECB and NFP?

Great post CanOz.

These kinds of questions are a great way to gain some perspective on how things are moving

 

[CanOz]

Actually the ECB only reports 8 times per year on rate decisions....so the two days is mostly waiting for the NFP. There 3 occasions, including this week that the ECB reported the day before the NFP report.

I'm trying to look for trend days after they report, but not seeing much in it yet.