Modelling the Drawdowns to Improve Performance
- Thread starter Rapunzel
- Start date
Start with the premise - risk is of an event happening, or not.
What risk is not is losing X numbers of dollars.
I'll try when I have time to find pertinent part in video.
hmmm....not quite. Maybe for event driven trading or something similar, not what I do. The entire premise is to gain statistical edge over a large number of trades, by looking for market properties (ideas) that tilt the probabilities.
You wouldn't be able to even conceive a problem (trade) to investigate without having some strict criteria like r:r with which to measure the effect (or not) of your idea on the outcomes.
Last edited:
I think the problem is a different approach. If you trade the Price Action Intraday using Technical Analysis then it would not make sense to have a fixed R:R.
I feel like you need to have access to the order book of your broker, to evaluate the real price action, in-order to make TA work properly, otherwise you're just guessing. So it's probably not appropriate as a bread-and-butter approach to trading unless you have the science to back it up.
Maybe trading fundamental news and things like that are probably more likely to work without seeing the orderbook but then by their very nature arise 'randomly'/ situationally.
I feel like you need to have access to the order book of your broker, to evaluate the real price action, in-order to make TA work properly, otherwise you're just guessing. So it's probably not appropriate as a bread-and-butter approach to trading unless you have the science to back it up.
Maybe trading fundamental news and things like that are probably more likely to work without seeing the orderbook but then by their very nature arise 'randomly'/ situationally.
This painting by Edvard Munch epitomize trading.
It's a scary journey. With a bit of imagination, observing the number of lows and it's distribution for a period of about 2 months, it can be seen old lows are taken out with new ones. It is when the new lows become scarce that is where opportunities present itself. Why not just look at Q4 2008 - Q1 2009 (global financial crisis), and Q1 2020(pandemic) and have a feel for yourself.
It's a scary journey. With a bit of imagination, observing the number of lows and it's distribution for a period of about 2 months, it can be seen old lows are taken out with new ones. It is when the new lows become scarce that is where opportunities present itself. Why not just look at Q4 2008 - Q1 2009 (global financial crisis), and Q1 2020(pandemic) and have a feel for yourself.
Yes, we have taken a snapshot of those times you mention. You have been found wanting, because these are not those times. We don't need to try to equate The Scream with trading in 2022. It is a different world.
Get a grip.
Hi thanks for this earlier post. I was thinking that a good property might be one that uses the standard deviation as a measure of 'down-side'/other-side risk.
Something similar to Sharpe Ratio might a way to do it. If the sharpe ratio measures the risk-adjusted return (https://www.investopedia.com/terms/s/sharperatio.asp) by taking the return and dividing it by the standard deviation of the return (volatility), then you could take that concept and apply it to price movement: the price movement in a direction divided the volatility of the price movements.
Or better still; the concept of Sortino; the price movement up versus the volatility of downward price movement and the price movement down, versus the volatility of upward price movement.
The progression of that property over time might be a good way to predict drawdowns in a trade at a particular point in time, or at least help quantify the risk.
Okay.
Just had a quick look. The property/indicator is for 20 bars looking at the buys as example is
average price change for bars that closed up / st_deviation price change for bars that closed down
quick test EURUSD 2004 - 2017 H1.
the value ranges from 0 to 12. As it increases the average drawdown on the trades decreases. The relationship is more pronounced on the maximum drawdown on the trades. So there is something there, need to look at it in more detail.
average price change for bars that closed up / st_deviation price change for bars that closed down
quick test EURUSD 2004 - 2017 H1.
the value ranges from 0 to 12. As it increases the average drawdown on the trades decreases. The relationship is more pronounced on the maximum drawdown on the trades. So there is something there, need to look at it in more detail.
Last edited:
Hi incidentally, i've just been playing around and there may be a price action indicator in it. If you take a short number of bars (like 10) and plot 2 graphs:
(1) buy downside risk = average price change up in 10 / st dev price change down in 10
(2) sell upside risk = -1* average price change down in 10 / st dev price change up in 10
Looking at the graphs, there seems to be a a good indicator of future upward price movement after a cross over of (1) over (2) and vice-versa.
(1) buy downside risk = average price change up in 10 / st dev price change down in 10
(2) sell upside risk = -1* average price change down in 10 / st dev price change up in 10
Looking at the graphs, there seems to be a a good indicator of future upward price movement after a cross over of (1) over (2) and vice-versa.