MultiCharts Standard vs MultiCharts.NET

Discussion in 'Trading Software' started by Dan_public, Jul 22, 2017.

  1. Hi. After more than a decade of not trading, I'm getting back into it. And I need help on choosing the right charting/analysis/execution software for swing trading. Some history...

    I started day trading in the late 1990's. For the next 10 years, I used several brokers and software charting/strategy software packages including IB, TDAmeritrade (TOS), AmiBroker, and TradeStation. For a variety of reasons, now they will only provide a partial solution.

    The new software has to have good charting and back testing, have an automatic trading interface, run reasonably fast, and be very stable. IB supports a decent API and has a good commission structure, so they are fine for the broker part right now. For the charting, back testing, and automation part...

    After much thought, and testing the trial standard and .NET versions of MultiCharts (plus IQFeed), they seem to fit my needs. Maybe.

    I'd appreciate your feedback about MultiCharts in general and the two versions in particular. My specific concerns relate to which version of MultiCharts is better to implement my strategies - MC/EasyLanguage or MC.NET?

    In general, my strategies involve scanning 200 to 400 instruments daily looking for certain short term price/volume divergences over the last 20 days or so. But here's where it gets difficult:

    1. The starting point is a 200-400 base symbol list generated automatically or at least semi-automatically every day. The list would be created by MultiCharts. Or maybe by something like FinViz (www.finviz.com) and then imported into a MultiCharts watchlist.

    2. The MC watchlist would then be screened using standard and custom MC indicators looking for very specific patterns. Since the specific pattern is somewhat rare, I expect only 10-20 symbols per week would fit the pattern. This watchlist scanning should NOT require charts.

    3. The final 10-20 symbols would be loaded into another watchlist for final scrubbing, chart review, and automatic trade execution. ​

    The key question is which version of MultiCharts would work best...

    The MC/EasyLanguage has lots of well-known benefits, making it very attractive for non-programmers. However, I am a C# and database programmer who builds analytical systems professionally. So using MC.NET is not an issue.

    On the other hand, MC/EasyLanguage seems easier to code and has lots of readily available indicators and signals in EasyLanguage. Originally, I was swayed toward MC/EasyLanguage.

    Then I found two interesting posts in the Ampfutures forum comparing MC/EasyLanguage and MC.NET: https://forum.ampfutures.com/forum/...s-platform/160-multicharts-vs-multicharts-net . The posts said that several MC.NET features are NOT available in
    MC/EasyLanguage:

    • Access to all information in Order and Position Tracker from scripts
    • Access all data from Scripts for instruments that are not even charted
    • Access to the list of symbols in the database from studies
    • Ability to use the third-party databases like SQL Server and Mongo DB
    • Access to the status of orders, positions, accounts, logs from the script
    Given my strategies above, a lot of processing involves batch database work. The only need for a chart should be in the last part of step 3.

    Please help me. Is all of this correct? Does only MC.NET have these features? Are there workarounds in MC/EasyLanguage? Any overall advice?

    Thanks in advance for any feedback.

    Best regards,

    Dan.
     
  2. WeToddDid2

    WeToddDid2

    My personal experience is that easylanguage sucks. Years ago I was new to coding and decided to learn easylanguage for Tradestation to code several different strategies. After learning the language and working for months coding the strategies, I ran into several limitations of EL. EL was the first programming language that I learned. I am still pissed off about wasting so much time.

    Frankly, I am having a hard time believing that your question above is genuine. I find it hard to believe that a C# coder would even consider a limited language like easylanguage. Just seems totally illogical. The above mistake I made was because I was green. I can't imagine an experienced coder making the same mistake. Honestly, seems more likely that you are doing marketing research for Multicharts.

    Personally, I would go with none of the above in your list of platforms.

    There are several other C# options that you may want to consider. Not sure if the below platforms meet all your needs. There are several other institutional level options but expensive.
    IMHO, if i were to learn a new language at this time for trading it would by python for sure but C# is arguably the best language for writing trading algos. Some python options below.

     
    Last edited: Jul 23, 2017
  3. bobvago

    bobvago

    I agree. Easylanguage sucks. It is limited in many ways.
     
  4. MattZ

    MattZ Sponsor

    Dan, I truly think that you should test your strategies on both MCs and determine what works best for you. ( We can provide MC.NET for free if it helps).
    That way you can have the confidence implementing a long term strategy.
    Since you want to see all your working orders, I recommend using the Rithmic data feed for your MCs.
    First, you would be able to get unfiltered data as delivered by the exchange and also have access to their own platform (Rithmic Trader) that is capable of tracking all your orders both working, filled and cancelled.
     
  5. WeToddDid2,

    Many thanks for the feedback.

    Regarding EasyLanguage, please understand that I have very limited understanding of EL. So I have no reference for comparison. That said, I was pretty sure that C# was the more robust option. However...

    One reason I even considered MC/EL is that I love to write code. That's a double-edged sword. If I used MC/EL I think I'd focus more on strategies and algorithms. If I choose MC.NET, I'd probably write lots of code. That may be more enjoyable, but spending time writing lots of code may not make me money. (I did exactly that with AmiBroker in the early to mid 2000's.) Now, my goal now is to make money.

    In any case, I'm not sure if MC.NET is the right choice either. One critical part of my plan is requirement to import a symbol text file into a watchlist for processing. I spent some time this evening trying to import a list of symbols into MC.NET. I couldn't find any obvious way to do that. I'm going to call their tech support tomorrow to get more info.

    Before I installed MC, I installed NinjaCharts. Like MC.NET, I couldn't find any way to do that even after a fair amount of work. I'm going to call them tomorrow for feedback.

    Thanks for the links to IQBroker and the others. ...

    - IQBroker looked very interesting initially, but after digging deeper, there are few things that concern me. The documentation is very sparse with no scripting examples that I could find. The IQBroker forum has 22 posts total and most are from the Admin. The most worrying thing is that there is no one to provide phone support, even sales support. The only way to communicate with the company is to send an email or post in the support forum. And finally... I could not figure out a way to import a list of symbols.

    The strange thing about IQBroker is that the software looks good overall, felt smooth and solid, and connected to IQFeed immediately - no hiccups. And their website is fairly sophisticated. Overall, I don't understand them.

    - QuantConnect looks interesting but confusing. In some parts of the web page it implies that the data feeds are free. In others, the historical and real-time data is in the thousands of dollars. I have to study them more.

    - quantiacs and quantopian are a different model for me. I don't feel comfortable with that.

    Again, many thanks for the good feedback. If you see anything I've missed, please ping me on it.

    Best regards,

    Dan.
     
  6. WeToddDid2

    WeToddDid2

    Dan,

    Thank you for your feedback.

    Can you please elaborate with respect to quantiacs and quantopian being a different model?
     
  7. WeToddDid2,

    Hi. Perhaps I don't understand their model, but it appears that your trading strategies are passed to expert traders with large accounts and then you share in the profits. I can see a lot of holes with that concept, so I don't feel comfortable with that.

    Am I wrong? Possibly. Maybe event probably. I have a very strong negative reaction to anything that smells like hype. It's as simple as that.

    Best regards,

    Dan.
     
  8. JSOP

    JSOP

    Dan, with your strategy, you would have no choice but to use MC.Net UNLESS you want to put your 100-200 symbols onto the charts first, then you would be able to use both EasyLanguage and .Net languages (C# and VB) but even then you can only put max. of 50 symbols onto the charts as subcharts so EasyLanguage still won't be an option to you.

    To me the choice between EasyLanguage and MultiCharts.Net is an issue of learning propensity/ability in programming. EasyLanguage is as what its names says, a REALLY EASY language that is simple and FAST to learn and it's more than adequate for automatic trading purposes but it won't allow you to code everything that you want your strategy to do UNLESS you are willing to modify your strategy in a way that it doesn't require everything to be automated. MC.Net programming, on the other hand is a LOT more in that it provides more flexibility and capabilities because it's a more machine-level language and it IS able to do EVERYTHING what you want your strategy to do according to how you described it BUT its learning curve is REALLY steep especially if you have little programming experience/knowledge with particularly OO (which I feel is the most useless and inefficient and stupid concept ever invented) programming. It's a good thing that C# is not pure OO like the useless and slow Java (IB is written in Java. Need I say more?) and provides some room for functioning programming which is what programming should be.

    Good luck!
     
    Last edited: Jul 27, 2017
  9. JSOP,

    Thanks for your feedback. I agree on all points, especially OO. Some aspects of OO are useful, but it can be unreadable past a certain point.

    I didn't know that IB was written in Java. That explains a few things.

    Regards,

    Dan.
     
  10. MarkBrown

    MarkBrown

    Dan,

    Like you I have had many platforms. I still use TS just because I have decades invested in the language. I also have MC because it shares EL studies and it hooks up to many feeds that TS does not including some really good feeds that are premium.

    I also have AmiBroker and think it's the best long term investment I have for running sophisticated multi-system multi-symbol, well I look at it as the future.

    If I were you and wanted some control over your systems and studies, MC that is EL compatible would be your best bet just because you can do some if not all your "own thing".

    Mark
     
    #10     Jul 31, 2017