Trading volatility - a possibly interesting exercise | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Price Action Exercise: What Happened Next? 49 replies

Exercise for Newbs using VHands 13 replies

An exercise in frustration 5 replies

Money Management exercise 5 replies

Trading volatility - a possibly interesting exercise

  • #21
  • Dec 23, 2010 9:00am Dec 23, 2010 9:00am
Quoting Scotty B
For example, suppose you knew that the market would have a probability of 94-96% of having at least a 70 pip range today. Could you build a trading system on that objective information?
Easy, after 50 pips in one direction, after the middle of the US session and considering that there aren't news anymore, I'd go for 20 pips TP with 40 pips SL. I'd win at least in 80% of the times.
  • #22
  • Dec 23, 2010 10:12am Dec 23, 2010 10:12am
Quoting Sauron
Easy, after 50 pips in one direction, after the middle of the US session and considering that there aren't news anymore, I'd go for 20 pips TP with 40 pips SL. I'd win at least in 80% of the times.
You'd be surprised how many late day spike reversals there are. This does not have to be a problem though, you just need to know the nature of daily reversals, i.e. on average, how many times does price reverse to break its current high or low? So, without getting too creative, it's likely not enough information just to know today's range-- we need more data to hang our hats on.

This is one of the biggest lessons I've learned and or realized through all my OFA studies. Every thing on a chart we use as a "signal," has it's own unique probability, and all we do when we take a trade on such a signal is to buy into that probability. It's impossible to escape the statistics. The first priority when looking at any trading system is to look at these numbers. Duh, right?

Now, I'm probably opening up a can of worms here, but when you start with objective data, could you build a controlled martingale strategy when you operate with the odds on your side? What type of data would you need? This is just one idea, but the options are many when you finally have something solid to work with.
  • #23
  • Dec 23, 2010 12:44pm Dec 23, 2010 12:44pm
You can't trick the math. equation even if the spread is 0

In reality the twice paid spread will kill you.
  • #24
  • Dec 24, 2010 3:43am Dec 24, 2010 3:43am
Quoting Scotty B
You'd be surprised how many late day spike reversals there are.
There aren't so many without news during the asian or aussie session and I factored them in. One can move the market with lower volume during this sessions but without a very good reason it will not happen.

Quoting Scotty B
It's impossible to escape the statistics. The first priority when looking at any trading system is to look at these numbers. Duh, right?
You're still a little bit under the spell of DS and his high win ratio. I don't have a problem taking losses as long as the winnings are bigger. If the odds are on my side I don't care that I don't know what will happen on the next trade. I've just quoted somehow from Douglas, shame on me.

Quoting Scotty B
Now, I'm probably opening up a can of worms here, but when you start with objective data, could you build a controlled martingale strategy when you operate with the odds on your side?
If the odds are on your side you don't need the extra risk of a martingale approach. It only creates a safe feeling that you don't lose till you lose big.
  • #25
  • Dec 24, 2010 6:21am Dec 24, 2010 6:21am
Quoting Sauron
There aren't so many without news during the asian or aussie session and I factored them in. One can move the market with lower volume during this sessions but without a very good reason it will not happen.


You're still a little bit under the spell of DS and his high win ratio. I don't have a problem taking losses as long as the winnings are bigger. If the odds are on my side I don't care that I don't know what will happen on the next trade. I've just quoted somehow from Douglas, shame on me.


If the odds are on your side you don't need...
It's a percentage game when we are talking about daily reversals. If you knew for example that 90% of the time there are at least 3 reversals, 80% of the time there are 4, 70% there are 5, 50% there are 6 ect.. I made those stats up by the way. Now, automatically there are several options in how one could use those numbers. In referance to trading to a daily range target, you could begin trading out toward the target on the 5th reversal break and limit your martingale to 5 position reversals, thus avoiding the "big loss," hence the control aspect of the martingale.

My point isn't about how great a martingale approach is, just that when you know the odds your toolbox gets much bigger.

We are in agreement about the futility in worrying about the outcome of the next trade. It's over a series of trades that the profits emerge.
  • #26
  • Oct 25, 2017 5:37am Oct 25, 2017 5:37am
The thought is to always be in the market, entering/leaving positions every minute of every day, while never picking a course. You would just exchange unpredictability. This would obviously just work in a market that has fair/great intraday instability and doesn't slant much in the more drawn out term. Patterns murder it. It would work phenomenally in ranges. You would dependably be conveying drawdown, however suppose value moves to what will end up being the external edge of the range. You've collected a decent unrealised drawdown, and loads of acknowledged pips. Value at that point achieves what looking back is the edge of the range, and turns around.
  • #27
  • Last Post: Nov 3, 2017 11:15am Nov 3, 2017 11:15am
Quoting Scotty B
Instead approaching the market with your idea, try to approach some aspect of the market that you can quantify, then apply your idea--namely, trading without relying on an subjective prediction about the direction of the market. For example, suppose you knew that the market would have a probability of 94-96% of having at least a 70 pip range today. Could you build a trading system on that objective information?
Hey, @Scotty B, I just saw this now. What you mentioned above is what I've been searching for since last year. I'm looking for a way to accurately predict market volatility. If I know with a probability of 94-96% that the market would have at least a 70 pip range today (and I get that information yesterday), I would be rich in 6 months!
Please if you know any method of predicting volatility like that, let me know. I have tried ATR but it works partially - not very good for capitalising on breakouts. I have perfected a system that can help me profit EVERYDAY if I knew tomorrows range with a high level of certainty/probability ( at least 95%).

I'm looking into Volatility Modelling using GARCH and Stochastic Processes modelling. I'm still not very good in Maths despite having an economics background.
0 traders viewing now