News Avoidance | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Optimization, statistics, and avoidance of curve fitting 21 replies

Fake-out avoidance & targets 5 replies

Focus on loss avoidance strategies 2 replies

Stock Market Crash Avoidance 2 replies

Attachments: News Avoidance
Exit Attachments

News Avoidance

  • #21
  • Jul 13, 2017 10:53am Jul 13, 2017 10:53am
Quoting HeyYou
{quote} That's what I think as well.. avoid "curve fitting", invest money that you can afford to lose.. have faith. BUT, wait a sec, do you mean testing 10,000 setups on a single instrument ? gee that seems impossible to me even over a 20 years period on a very volatile instrument.
No not a single instrument, but the more the events better, it's all about deviation from the mean, take a 0 ev proposition the classic coin toss, we all know given a fair coin that there is an equal chance of either heads or tails coming up on a given toss, and even on say 10 tosses it is not that unusual to see 10 heads or 10 tails, the more tosses that are performed the closer the percentage of heads or tails gets to 50%, that's all there is to it the smaller the sample the more likely that there has been luck involved because of standard deviation from the mean, the larger the sample the more likely that the true edge or lack of it is being represented.

Worth reading about expected value and standard deviation if you aren't already knowledgable about those concepts, the other big thing is setting unit size for that I again use lot and lots of data to calculate the return over maximum drawdown, from this I can set my unit size based upon my personal risk appetite, which is for me a maximum drawdown from any given peak in the region of 15%. Look up RoMad and gamblers ruin, again assuming I am not teaching grandmothers to suck eggs here Investopedia is a very useful resource.

The last thing in my egg sucking tuition is confirmation bias, I recommend always using a computer to run your back tests and also to do your execution, humans are very susceptible to seeing what they want to see, computers do not have an emotional attachment to a given thesis.
I live entirely from the growth of my personal trading account
  • #22
  • Jul 13, 2017 11:09am Jul 13, 2017 11:09am
Quoting Madinvestor
I am starting this thread to get the collective opinions of those of you who avoid trading around news. I myself avoid trading 15-30 mins before the news. I make sure to only avoid high impact news events that affect my chosen pair aka the ones in red on the FX calendar and disregard all the medium to low impact events. What do you do? Do you avoid all news or just high impact? Just curious....
Everyone does what he does best.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_3.png Size: 30 KB

https://www.forexfactory.com/showthr...0.png?v=15.1">
Only the price on the chart can show the entrance to the deal...
  • #23
  • Jul 14, 2017 5:37am Jul 14, 2017 5:37am
Quoting GuyBaker
{quote} No not a single instrument, but the more the events better, it's all about deviation from the mean, take a 0 ev proposition the classic coin toss, we all know given a fair coin that there is an equal chance of either heads or tails coming up on a given toss, and even on say 10 tosses it is not that unusual to see 10 heads or 10 tails, the more tosses that are performed the closer the percentage of heads or tails gets to 50%, that's all there is to it the smaller the sample the more likely that there has been luck involved because of standard...

reeeeally ? reeeeally ?

the problem is that you can "curve fit" your strategy to work under certain conditions and yes you can be profitable after 1000 setups, but as the market changes your system will crash. I found scalping strategies that survived 2 years with amazing returns.. to then crash. that's why I think a solid strategy should be tested on several market conditions, rather than relying just on the number of setups.


Quoting GuyBaker
I recommend always using a computer to run your back tests and also to do your execution, humans are very susceptible to seeing what they want to see, computers do not have an emotional attachment to a given thesis.

not so easy! the ego, which is another funny thing in human beings will say "what!? I'm always objective in my analysis...", here's a brand new excuse NOT TO use EAs.
I think what separates a successful trader from a loser IS the lack of confirmation bias, for MOST OF THE TIME in his life.

P.s. sorry ppl for my English.. you grammar nazis
  • #24
  • Jul 14, 2017 5:55am Jul 14, 2017 5:55am
Quoting HeyYou
{quote} reeeeally ? reeeeally ? the problem is that you can "curve fit" your strategy to work under certain conditions and yes you can be profitable after 1000 setups, but as the market changes your system will crash. I found scalping strategies that survived 2 years with amazing returns.. to then crash. that's why I think a solid strategy should be tested on several market conditions, rather than relying just on the number of setups. {quote} not so easy! the ego, which is another funny thing in human beings will say "what!? I'm always objective...

Curve fitting is easy to avoid, it just requires awareness that it exists.

I am not a fan of discretionary trading so for me the easiest way to avoid confirmation bias is to use my computer for execution and testing, this also frees my time up so I can sit on the beach I am a software engineer by training though, so I recognise it mis easier for me than many others.

Yes true various market conditions, I tested my Binary system all the way back to 2008 on a one minute, 2, minute, 5 minute and 20 minute charts on multiple pairs, my OTC Forex system back to 2003, all I am saying is the more testing the better, there are many people who try something for a day and decide, hey it's working....
I live entirely from the growth of my personal trading account
  • #25
  • Jul 14, 2017 6:03am Jul 14, 2017 6:03am
Quoting GuyBaker
{quote} Curve fitting is easy to avoid, it just requires awareness that it exists. I am not a fan of discretionary trading so for me the easiest way to avoid confirmation bias is to use my computer for execution and testing, this also frees my time up so I can sit on the beach I am a software engineer by training though, so I recognise it mis easier for me than many others. Yes true various market conditions, I tested my Binary system all the way back to 2008 on a one minute, 2, minute, 5 minute and 20 minute charts on multiple pairs, my...
fact is, you won't get rid of "confirmation bias" in any way IMO... naive people will always be naive.. just look at these forums, you have people who signed up 10 years ago, still have no clue what they're doing.

nobody is perfect and sometimes we can afford to waste money, or out of generosity.. we so love brokers .lol
  • #26
  • Jul 14, 2017 6:19am Jul 14, 2017 6:19am
Quoting GuyBaker
{quote} Curve fitting is easy to avoid, it just requires awareness that it exists. I am not a fan of discretionary trading so for me the easiest way to avoid confirmation bias is to use my computer for execution and testing, this also frees my time up so I can sit on the beach I am a software engineer by training though, so I recognise it mis easier for me than many others. Yes true various market conditions, I tested my Binary system all the way back to 2008 on a one minute, 2, minute, 5 minute and 20 minute charts on multiple pairs, my...
That makes more sense now. If I decide to do any historical testing, it all has to be done by hand and I can only use so many instances because of that. I am a discretionary trader, so my strategy cannot be programmed. This is why I have historically avoided trading the news in my strategy. It just brings to much uncertainty and would be gambling for me at this point.
  • #27
  • Jul 14, 2017 6:20am Jul 14, 2017 6:20am
Quoting HeyYou
{quote}.. just look at these forums, you have people who signed up 10 years ago, still have no clue what they're doing.
Lol, yeah that's sad, but true.
  • #28
  • Jul 14, 2017 6:37am Jul 14, 2017 6:37am
Quoting HeyYou
{quote} fact is, you won't get rid of "confirmation bias" in any way IMO... naive people will always be naive.. just look at these forums, you have people who signed up 10 years ago, still have no clue what they're doing. nobody is perfect and sometimes we can afford to waste money, or out of generosity.. we so love brokers .lol

The missing factor is being able to look at oneself objectively, unless a person can do that he is lost, and not just in trading, although trading makes our personal flaws abundantly clear, I tested my manual back testing skills and found them lacking, I also tested my discretionary trading skills and found them abysmal, so after a long time repeating the same mistakes I finally went back to what I had always known, computers are my forte, it seems I had to take the path to ruin before I could find my way to success, a kind of self sabotage I guess, this shit is hard, and then one day it becomes easy.
I live entirely from the growth of my personal trading account
  • #29
  • Jul 14, 2017 6:54am Jul 14, 2017 6:54am
Quoting Madinvestor
{quote} That makes more sense now. If I decide to do any historical testing, it all has to be done by hand and I can only use so many instances because of that. I am a discretionary trader, so my strategy cannot be programmed. This is why I have historically avoided trading the news in my strategy. It just brings to much uncertainty and would be gambling for me at this point.
I wish you all the best mad.


p.s. I'm leaving this discussion lol , we're a bit OT
  • #30
  • Jul 14, 2017 7:27am Jul 14, 2017 7:27am
Quoting HeyYou
{quote} I wish you all the best mad. p.s. I'm leaving this discussion lol , we're a bit OT
Likewise! Thanks for sharing your thoughts.
  • #31
  • Jul 14, 2017 10:41pm Jul 14, 2017 10:41pm
Take for instance intraday traders who have small sl say 5-20pips. A red news can cause slippage in the order of 30-100+ pips depending on the nature of the news. This will cause significant gain/loss in an instant at the news release point. That's why most intraday traders who place trades based off TA don't hold open positions going into red news releases - that's not their game. Ofc immediately after that moment when this slippage risk is over the trader can be back trading the market as usual.

However, if the open position is sitting on positive gains that might withstand the potential negative slippage then the trader may take the risk to not close his trade only if he expects the news release to favour his trade. The usual case is to open positions early to allow for a profit buffer then ride the news release spikes towards the profit target - this do require prior FA analysis in the trade plan.
  • #32
  • Last Post: Jul 15, 2017 3:38am Jul 15, 2017 3:38am
Quoting skyway
Take for instance intraday traders who have small sl say 5-20pips. A red news can cause slippage in the order of 30-100+ pips depending on the nature of the news. This will cause significant gain/loss in an instant at the news release point. That's why most intraday traders who place trades based off TA don't hold open positions going into red news releases - that's not their game. Ofc immediately after that moment when this slippage risk is over the trader can be back trading the market as usual. However, if the open position is sitting on positive...
I don't think you used them slippage in the right context. I think what you meant was the volatility but I definitely agree with the overall point. Not to be picky about terminology, (someone correct me if you think I'm wrong) but from what I've always known slippage refers to different between your expected execution price vs the actual full price. It can only happen on market orders as limit orders world simply not fill if the price was larger or smaller than the order was for. So if you put on a market order during news for price .6778 and spread was 2 pips, you would expect your actual entry price to be .6780. However if it is very volatile at the time due to a news event, the price is moving so fast that maybe you get an actual full of .6785. that means you got slippage of 5 pips on your order.
0 traders viewing now