Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
With these 14 years of 30m FDAX data you can probably find a nice edge.
There are many options, and an important question is: What’s important?
In my current process I normally filter the data on weekday, prior day OC and HL, 2 days ago OC.
Then I’m adding gap Y/N, where price is at 9:30 (inside/below/above prior day HL) and the direction/size of IB (9:00-9:30 bar).
In case not many historical occurrences (with these same conditions) remain after filtering (which is often the case when there is a gap), I’m releasing filter prior day HL, 2 days ago OC and sometimes prior day OC.
I’m using this 14 years of data to have enough historical occurrences. To me, it doesn’t make much sense to work with 14 years of data for day-trades. Right??
I give great importance to the IB direction/size. Also to the weekday filter, to the gap and whether it closed, and where price is at 9:30.
Why? Basically because gives importance to these in his journal. And it makes sense to me.
I am filtering on conditions that are similar to current day’s conditions, to then see which trades (trading direction/time/SL) would have brought good results in the past.
It makes sense to me, but I guess that’s just based on assumptions.
If (for instance) I wouldn’t use a filter on the weekday, the system would give me roughly 5 times more occurrences, so I could for instance filter on a smaller range for IB, and the statistics might become more reliable.
What do you guys think about all this?
How we can use this data/tool to get usable statistics?
If there is any (30m FDAX) statistic you would like to see, I’ll be happy to share it.
To prepare for today I entered the values for prior day OHLC and the day before OC.
Also I filtered on day 4/Thursday. And a filter on 'up gap'.
GREEN IB
When I add a filter to a green IB (9:00-9:30 bar), the output looks like this:
RED IB
When I add a filter to a red IB (9:00-9:30 bar), the output looks like this:
When there is a gap, not many historical occurrences are available.
Still I have to filter whether the gap is closed or not, where price is at 9:30, and the IB size.
So I'll probably have to release filters for prior day and the day before.
GSIs1 definitely thinks the day of the week is important. He almost always (but definitely not always) goes short on Wednesdays for example.
You might be able to combine some days together (Tues and Thurs?) but not others?
Not a complete thought but to trigger me to more fully explain later: Weight the data and combine until a sufficient sample size has been achieved i.e. Exact match data is weighted higher than partial-match data and/or voting agents.
Gap: up gap 38pts
Gap closed during IB: no
Position price at 9:30: above prior day HL
IB direction: long
These are the results using filters:
Thursday
Up gap not closed
IB range
As there are 28 occurrences, following the flow chart posted a few days ago, I'm adding prior day oc to the filter:
Still 22 occurrences, so I'm adding the OC of 2 days ago:
Now the long trade 10:30-11:00 from the first 2 snap shots are gone from the top of the list due to a low win%:
So yeah, while the long trade 10:30-11:00 looks good at first, it doesn't once the OC of 2 days ago is added to the filters.
Following the flow chart, thus no trade.
Thanks
Deetee
(SumR translates to a relative (to current FDAX price) result of X points per trade
(Relative trade result = trade result in pts / 9am open price * 10000, and the SumR is the sum of the relative trade results))
NexusFI website is down, so can’t post in real-time.
FDAX
Preparation:
For the long IB, the results look good. 74% win% based on 27 trades and a SumR of 312.
For the short IB:
After IB completed: Gap: Yes, down only 2 pts, so not counting this as a gap.
Gap closed during IB: yes, in seconds after opening
Position price at 9:30: below prior day HL
IB direction: short
I’m excluding the up gap, and gap not closed days.
Gap: No
Gap closed during IB: x
Position price at 9:30: inside prior day HL
IB direction: short 3 pts
IB HL 102 pts (down) and fully came back.
I’m not using a filter on the IB HL extreme values, but would have preferred to be able to today.
Anyway, I’m following the flow chart. It gives nice output with an 87,5% win%.
Sounds like a sure winner, IF we could only rely on the system
DAX long (paper) trade
Entry 10:30
Exit 13:00
SL 35 pts
Result 35 pts loss
Historical results with this setup: See snapshot
Let’s see
(SumR translates to a relative (to current FDAX price) result of X points per trade
(Relative trade result = trade result in pts / 9am open price * 10000, and the SumR is the sum of the relative trade results))
To prepare for today I entered the values for prior day OHLC and the day before OC.
Also I filtered on day 2/Tuesday. And a filter on 'no down gap' as current price is about 200 pts higher then yesterday’s low.
GREEN IB
When I add a filter to a green IB (9:00-9:30 bar), the output looks like this:
RED IB
When I add a filter to a red IB (9:00-9:30 bar), the output looks like this:
Not very exciting yet.
Let's see how things look around 9:30 CET.
I want to try this system/tool, but without filtering on the prior days (unless prior day OC or HL had extreme values).
Meaning, focus on the IB direction/size, the gap and whether it closed, and where price is at 9:30.
Not sure yet about the weekday filter, but could run a parallel test.
Not filtering on the prior days will give more occurences, and allows me to narrow the IB OC range that I have been using.
I'll continue with what I was doing, but am going to check the results without using the prior days.
Will need to create more narrow OB OC ranges, that will take me some time.