NexusFi: Find Your Edge


Home Menu

 





Support and Resistance on Crude Oil


Discussion in Traders Hideout

Updated
      Top Posters
    1. looks_one Crow with 5 posts (7 thanks)
    2. looks_two xplorer with 3 posts (6 thanks)
    3. looks_3 Quick Summary with 1 posts (0 thanks)
    4. looks_4 bobwest with 1 posts (3 thanks)
    1. trending_up 1,908 views
    2. thumb_up 16 thanks given
    3. group 2 followers
    1. forum 9 posts
    2. attach_file 4 attachments




 
Search this Thread
  #1 (permalink)
 Crow 
Las Vegas, Nevada
 
Experience: None
Platform: MultiCharts
Posts: 52 since Apr 2010
Thanks Given: 20
Thanks Received: 19

I have never traded futures, but I have spent a lot of time looking at crude oil futures. I wrote some code to automatically identify significant support and resistance levels because I want to objectively study any approach that I use and prove to myself that it works over time. When I am looking at charts, I have a tendency to see what I want to see so being able to put my ideas in code and prove that they are statistically significant saves me a lot of time. So I did that to identify and trade based on support and resistance levels on crude oil. I found a very small or negligible advantage compared to completely random entries. The problem is that too often, the support and resistance levels simply don't hold. Am I looking at the wrong instrument? Crude oil is supposed to move a lot more typically than equities or other futures. Is it the case that support and resistance just doesn't work as well for crude vs. S&P or something else?

Second question, does order flow really provide and edge in day trading? When I looked for other information to incorporate into my charts, the bid / ask is one thing to add, but I don't see anyone using it in their trading journals and when I google search it, I get all kinds of links to vendors and training classes...vendors with stuff to sell. I also see some pages where everyone loves it, no one has anything negative to say about it, but I see people use Fibonacci retracements a lot more frequently than order flow. Makes me think that if it was really good, I'd see it on every chart.

Finally, anyone here use Elliot Wave? Is there any value at all in studying it?

Started this thread Reply With Quote
Thanked by:

Can you help answer these questions
from other members on NexusFi?
What a Cool Draw/Markup Tool!
Traders Hideout
IB Gateway / IBKR (semi-) daily restart / login
Platforms and Indicators
Server CPUs and Multicharts 14 or 15 limits
MultiCharts
NinjaTrader indicator for last bar percentages
NinjaTrader
How to: Multi Time-Frame Volume Profile?
NinjaTrader
 
Best Threads (Most Thanked)
in the last 7 days on NexusFi
Fighting Addiction and Stops
17 thanks
Big Mike in Ecuador
6 thanks
Just another trading journal: PA, Wyckoff & Trends
6 thanks
Crossing the Abyss: An Adventure Guide by Snax
4 thanks
Want your NinjaTrader indicator created, free?
4 thanks
  #3 (permalink)
 
xplorer's Avatar
 xplorer 
London UK
Site Moderator
 
Experience: Beginner
Platform: CQG
Broker: S5
Trading: Futures
Posts: 6,088 since Sep 2015
Thanks Given: 15,688
Thanks Received: 15,861



Crow View Post
I have never traded futures, but I have spent a lot of time looking at crude oil futures. I wrote some code to automatically identify significant support and resistance levels because I want to objectively study any approach that I use and prove to myself that it works over time. When I am looking at charts, I have a tendency to see what I want to see so being able to put my ideas in code and prove that they are statistically significant saves me a lot of time. So I did that to identify and trade based on support and resistance levels on crude oil. I found a very small or negligible advantage compared to completely random entries. The problem is that too often, the support and resistance levels simply don't hold. Am I looking at the wrong instrument? Crude oil is supposed to move a lot more typically than equities or other futures. Is it the case that support and resistance just doesn't work as well for crude vs. S&P or something else?

Second question, does order flow really provide and edge in day trading? When I looked for other information to incorporate into my charts, the bid / ask is one thing to add, but I don't see anyone using it in their trading journals and when I google search it, I get all kinds of links to vendors and training classes...vendors with stuff to sell. I also see some pages where everyone loves it, no one has anything negative to say about it, but I see people use Fibonacci retracements a lot more frequently than order flow. Makes me think that if it was really good, I'd see it on every chart.

Finally, anyone here use Elliot Wave? Is there any value at all in studying it?

I can give you my take on the first two questions

Placing support and resistance lines is a skill. Trying to automate it is equivalent to trying to automate trading: the two activities require two separate skillsets. Also, you don't appear to mention timeframe. What timeframe are you placing s/r lines on, and what consequent risk/reward are you shooting for?

As for order flow, and whether it provides an edge in day trading, one could ask the same of virtually anything. Do fibonacci lines provide an edge? Do s/r lines? I am sure all of these could be debated and many people would sit on either side of the fence for each. I don't mean for this answer to sound flippant, I simply believe the secret is how well you know a certain tool. If you are starting now with order flow then you may not necessarily have an edge. But if you have used order flow for a long time then you may start seeing repetable patterns that, eventually, you should be able to trade with an edge, just like with any other trading tool.

My 2 cents.

Reply With Quote
Thanked by:
  #4 (permalink)
 Crow 
Las Vegas, Nevada
 
Experience: None
Platform: MultiCharts
Posts: 52 since Apr 2010
Thanks Given: 20
Thanks Received: 19

Thanks Xplorer. I'm using a volume chart where each bar is 330 contracts. No special significance of 330, I have tried larger volume bars such as 990 and found similar results. The smaller bars just show a bit more price action around the S/R levels and allow me to use smaller stops in an effort to keep my risk small. I have tried different risk:reward ratios, but for a 330 contract bar chart, I find that using a stop of 9 ticks against a target of at least 25 ticks away. I can't claim that is the optimal ratio, it's probably not, but my core problem is that support and resistance lines don't seem to be very reliable. Not sure if that's due to the instrument that I am trading or some other factors.

Started this thread Reply With Quote
Thanked by:
  #5 (permalink)
 
xplorer's Avatar
 xplorer 
London UK
Site Moderator
 
Experience: Beginner
Platform: CQG
Broker: S5
Trading: Futures
Posts: 6,088 since Sep 2015
Thanks Given: 15,688
Thanks Received: 15,861


Crow View Post
Thanks Xplorer. I'm using a volume chart where each bar is 330 contracts. No special significance of 330, I have tried larger volume bars such as 990 and found similar results. The smaller bars just show a bit more price action around the S/R levels and allow me to use smaller stops in an effort to keep my risk small. I have tried different risk:reward ratios, but for a 330 contract bar chart, I find that using a stop of 9 ticks against a target of at least 25 ticks away. I can't claim that is the optimal ratio, it's probably not, but my core problem is that support and resistance lines don't seem to be very reliable. Not sure if that's due to the instrument that I am trading or some other factors.

It comes down to what you mean by reliability. You're probably aware that support and resistance may have a different behaviour according to context, i.e. the trading day type and the way it develops. On ranging days you may found s/r lines to work more, whereas on trending days they may not necessarily hold. Especially on CL there are players that wait for certain s/r lines to be reached to then try and run them over and get others offside, so one needs to be mindful of that as well. Also, behaviour can change when liquidity dries up, such as for example in the minutes preceding and during releases such as EIA or API figures.

How did you arrive at 9:25 Risk:Reward?

Reply With Quote
Thanked by:
  #6 (permalink)
 Crow 
Las Vegas, Nevada
 
Experience: None
Platform: MultiCharts
Posts: 52 since Apr 2010
Thanks Given: 20
Thanks Received: 19


xplorer View Post
It comes down to what you mean by reliability. You're probably aware that support and resistance may have a different behaviour according to context, i.e. the trading day type and the way it develops. On ranging days you may found s/r lines to work more, whereas on trending days they may not necessarily hold. Especially on CL there are players that wait for certain s/r lines to be reached to then try and run them over and get others offside, so one needs to be mindful of that as well. Also, behaviour can change when liquidity dries up, such as for example in the minutes preceding and during releases such as EIA or API figures.

How did you arrive at 9:25 Risk:Reward?

Yes, I have noticed that support and resistance levels seem to work better when the price action is not in a strong trend. As for how I arrived at the 9:25 risk : reward ratio, well, what I did was basically use my automatic S/R level recognition code to identify S/R levels. I fixed the maximum width of the SR zone as 5 ticks (crude oil) and then pad it with 3 additional ticks of stop loss + 1 more tick of entry slack. Then I assume an ideal exit from that trade meaning that I allow the trade to run until it hits the stop and assume that I was magically able to get out at the best price (although to penalize allowing a huge stop loss, I subtract my stop loss from the ideal gain and I also cap the maximum gain at 100 ticks or actually 91 after subtracting the stop loss). That's an ideal situation that can never happen, but it allows me to isolate my entry strategy from my exit strategy. So with those parameters, the maximum ideal return would be maximized for about a 9 point stop loss with entry only in cases where there's no significant resistance (relative to the S/R level being traded off of) for at least 25 ticks. Again though, this is trading only off of support and resistance and considering nothing else -- no trend, moving average, or anything else to filter the trades.

Started this thread Reply With Quote
Thanked by:
  #7 (permalink)
 
xplorer's Avatar
 xplorer 
London UK
Site Moderator
 
Experience: Beginner
Platform: CQG
Broker: S5
Trading: Futures
Posts: 6,088 since Sep 2015
Thanks Given: 15,688
Thanks Received: 15,861


Crow View Post
Yes, I have noticed that support and resistance levels seem to work better when the price action is not in a strong trend. As for how I arrived at the 9:25 risk : reward ratio, well, what I did was basically use my automatic S/R level recognition code to identify S/R levels. I fixed the maximum width of the SR zone as 5 ticks (crude oil) and then pad it with 3 additional ticks of stop loss + 1 more tick of entry slack. Then I assume an ideal exit from that trade meaning that I allow the trade to run until it hits the stop and assume that I was magically able to get out at the best price (although to penalize allowing a huge stop loss, I subtract my stop loss from the ideal gain and I also cap the maximum gain at 100 ticks or actually 91 after subtracting the stop loss). That's an ideal situation that can never happen, but it allows me to isolate my entry strategy from my exit strategy. So with those parameters, the maximum ideal return would be maximized for about a 9 point stop loss with entry only in cases where there's no significant resistance (relative to the S/R level being traded off of) for at least 25 ticks. Again though, this is trading only off of support and resistance and considering nothing else -- no trend, moving average, or anything else to filter the trades.

I'm definitely no expert on algorithmic trading, but your reasoning sounds like something an algo/systematic trader would use to identify stop and target levels.

I've already stated my beliefs on skillsets of algo Vs discretionary trading - I will only add that, to me, s/r lines form naturally a part of discretionary trading. I have not seen them as part of a more systematic approach. That's not to say I have looked very hard though

Reply With Quote
Thanked by:
  #8 (permalink)
 Crow 
Las Vegas, Nevada
 
Experience: None
Platform: MultiCharts
Posts: 52 since Apr 2010
Thanks Given: 20
Thanks Received: 19

It's not easy to auto-generate support and resistance levels and even harder to generate non-horizontal trendlines, but it is possible. I'll try looking at the ES chart and see if the levels hold up better there. I suspect that they may.

Started this thread Reply With Quote
Thanked by:
  #9 (permalink)
 
bobwest's Avatar
 bobwest 
Western Florida
Site Moderator
 
Experience: Advanced
Platform: Sierra Chart
Trading: ES, YM
Frequency: Several times daily
Duration: Minutes
Posts: 8,174 since Jan 2013
Thanks Given: 57,981
Thanks Received: 26,386


Crow View Post
I have never traded futures, but I have spent a lot of time looking at crude oil futures. I wrote some code to automatically identify significant support and resistance levels because I want to objectively study any approach that I use and prove to myself that it works over time.
...
The problem is that too often, the support and resistance levels simply don't hold.


Crow View Post
It's not easy to auto-generate support and resistance levels and even harder to generate non-horizontal trendlines, but it is possible. I'll try looking at the ES chart and see if the levels hold up better there. I suspect that they may.

I think that if you place your support/resistance lines at horizontal levels based on turning points in price, you will find something like this:

1. There are just a whole lot of lines on your charts.

2. Most of the time, price just goes right through them. As has been stated, this happens less often during ranging periods than during trends, which is about what you would expect. (In a range, price has to stop somewhere, otherwise it's not a range....)

3. Even so, when price does stop or at least pause, it's often at (or very near) one of those lines or levels.

I would say that much the same is true when you use other ways to get your levels, such as prior highs/lows, floor pivots, VWAP, measured moves, and the many others that people use.

If #3 is a real thing (I think it is), and if #1 is also real (it definitely is), then this suggests that anticipated support/resistance may be helpful, but that if you use it alone you will get your head handed to you. I have seen a lot of trade journals here on NexusFi where the trader confidently makes a trade decision based on these levels (as well as on non-horizontal trend lines) and gets a horrible surprise. I've seen it when I do it, too.

In fact, something like "I expected price to (fill in the blank) at (x level), but it didn't" is a statement you read a lot.

So I think that if you do an automated backtest of levels formed by any method, you will find that, by themselves, they don't pan out too well.

There are also a number of traders on NexusFi who incorporate some set of levels in their trading and who do very well with them. But usually (I would say always, actually) they don't rely only on the levels by themselves, or just trustingly trade them automatically. They're part of some assessment of what price is doing at that point, but not generally the only thing.

Just my take on the question, anyway.

Bob.

Reply With Quote
Thanked by:
  #10 (permalink)
 Crow 
Las Vegas, Nevada
 
Experience: None
Platform: MultiCharts
Posts: 52 since Apr 2010
Thanks Given: 20
Thanks Received: 19


Here's an example of what looks like a good horizontal S/R level on crude oil 5-minute chart (July 1, 2018).



Many times, you can see price came very close to that level and did not break it. Only once did it break it, but only in 1 bar, the reversed. So let's see how that level works.



So price respected it very well on the first instance on that chart above. That would have been a very good short entry point. Low risk, good move. However, later, it violates it by 2 bars, then goes back under. Then price comes back, trades through it a while, but then on the next break, it moves through the level cleanly in one bar. Stays above it for a while, but then drop below cleanly on one bar and stays under for a while. You can see that when it does break through, it's usually cleanly, but then it can also trade through it for a while without really breaking through it by much. So if you tried to trade around that level, you would have to use a relatively large stop.

Let's look at the S&P e-mini:



This looks like a good S/R level. Scrolling forward:



As you can see, price respects the level, but unlike crude, it seems to stay on one side or the other of the level and when it breaches it, it either goes right through and continues on the other side or it will only break it by a few ticks and then reverse. Of course, it also doesn't tend to move as far as crude. But it seems easier to trade with S/R lines compared to crude. Has this been other people's experience as well? When I look at trading videos on youtube of crude oil, a lot of the time, the traders don't use S/R levels or trendlines. They seem to rely more on indicators.

Started this thread Reply With Quote
Thanked by:




Last Updated on November 10, 2018


© 2024 NexusFi™, s.a., All Rights Reserved.
Av Ricardo J. Alfaro, Century Tower, Panama City, Panama, Ph: +507 833-9432 (Panama and Intl), +1 888-312-3001 (USA and Canada)
All information is for educational use only and is not investment advice. There is a substantial risk of loss in trading commodity futures, stocks, options and foreign exchange products. Past performance is not indicative of future results.
About Us - Contact Us - Site Rules, Acceptable Use, and Terms and Conditions - Privacy Policy - Sitemap - Downloads - Top
no new posts