How I collect statistics
(my process to gain an edge)
a thread ๐งต
(my process to gain an edge)
a thread ๐งต
how do I come up with ideas to backtest etc?
whenever I am trading, I will at some point throughout the week have a question which pops up into my mind.
example:
how often do large wicks get filled and if they do when is the most likely case? the next day, 2 days, 3 days?
whenever I am trading, I will at some point throughout the week have a question which pops up into my mind.
example:
how often do large wicks get filled and if they do when is the most likely case? the next day, 2 days, 3 days?
setting parameters
what defines a large wick?
- 1%, 2%, 3%, 4%?
what time frame will I be calculating the wick size on?
- 15m, 1H, 4H etc?
set parameters for yourself to test the idea on
if these parameters come up unsuccessfully you can retest with different parameters
what defines a large wick?
- 1%, 2%, 3%, 4%?
what time frame will I be calculating the wick size on?
- 15m, 1H, 4H etc?
set parameters for yourself to test the idea on
if these parameters come up unsuccessfully you can retest with different parameters
collecting the data
for me, at minimum 1 year, most of the time I'm testing 2 years for my statistics
at this point it's now manual (if you don't have excel skills) go back 2 years and start manually checking for wicks
if you have some excel skills you can do the following...
for me, at minimum 1 year, most of the time I'm testing 2 years for my statistics
at this point it's now manual (if you don't have excel skills) go back 2 years and start manually checking for wicks
if you have some excel skills you can do the following...
collecting the data (2)
- extract tradingview data (higher plan needed)
data needed in this case:
open, high, low, close
use the data above to calculate a formula if either the high or low is (x%) distance from the close of the candle then print (โ ) in correct column
- extract tradingview data (higher plan needed)
data needed in this case:
open, high, low, close
use the data above to calculate a formula if either the high or low is (x%) distance from the close of the candle then print (โ ) in correct column
structuring the data
columns to have:
- date
- day of week
- session wick occurred in (A,L,N,C)
- upside or downside wick
- wick filled within 1 day
- within 2 days
- within 3 days
- more than 3 days
- if filled - session wick was filled in (optional)
example below ๐
columns to have:
- date
- day of week
- session wick occurred in (A,L,N,C)
- upside or downside wick
- wick filled within 1 day
- within 2 days
- within 3 days
- more than 3 days
- if filled - session wick was filled in (optional)
example below ๐
reviewing the data
once all columns have been filled in
(dating back to your select sample size)
select all the data put it into a pivot table
once all columns have been filled in
(dating back to your select sample size)
select all the data put it into a pivot table
Example
here is an example as to how I structured my pivot table when reviewing which session has the likelihood of putting in the daily high/low based on each individual day.
As well as the stats given to show me session probabilities for daily lows
here is an example as to how I structured my pivot table when reviewing which session has the likelihood of putting in the daily high/low based on each individual day.
As well as the stats given to show me session probabilities for daily lows
Hope you found this valuable as some form of kickstarter to start curating your own statistics.
I will be doing more deep dive into stuff like this going forward.
sub to the free newsletter (unreleased trade breakdowns) luckshury.beehiiv.com
I will be doing more deep dive into stuff like this going forward.
sub to the free newsletter (unreleased trade breakdowns) luckshury.beehiiv.com
Loading suggestions...