Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Best Renko Bar? - better vs median vs sbs vs wicked
No problem, had the same questions about 1 year ago, answers still fresh in my mind. Loved the Renko charts for too long!
You can download more market replay data (depending upon instrument) from NinjaTrader. There is a script somewhere on futures.io (formerly BMT) which automates the individual day downloads.
NT only save the BAR data, Open Close High Low and Volume. My understanding is they ware working on improving this in the new version. Product would be more useful with that addition. I like it for what it is, I have lots of gripes about it, but I manage to get around those issues.
NT's default Renko will not backtest correctly, because the Open of the bar is false. I believe Better Renko has solved that problem, but you can talk to other NT users to confirm that.
It is possible to create a secondary dataseries for execution. For example, create a 1-range bar for execution, and another dataseries for signal generation. But this method is for advanced programmers.
Nice idea on the script... I bet NT loves the loads on their servers . I've always wondered how reliable their MRP data really is. Seems like a better idea to save it with one's broker feed. Have you used their data okay? It'd be nice if Ninja would just convert saved historical data into MRP... no more "data" from those coders at NT.
I think the data is okay for what I need, may not be perfect but could be! I know when I tried recording it myself, I would always end up with gaps in the data due to Broker glitch or me need to reboot or testing something.
For Ninjatrader, I run a copy of it in a Chicago datacenter ('in the cloud') for maximum reliability and security with my live trades. I also record historical tick data and MRP data. I can export, compress and download both datasets to my desktop and work on the retail dataminig while everything runs live just fine. Executions are also quicker in the cloud, especially from Hawaii or some place. But just for how often Ninja crashes when 'working' it, it's worth it .
I share your enthusiasm for Renko charts, but had to say goodbye to them after constantly finding new ways they could 'trick' NT (and other platforms). Here are the issues that eventually led me to reluctantly set renko aside and trade other types of charts:
1. The apparent clarity of price action. The initial allure of Renko for me was the incredible clarity of price action compared to other charts (especially time charts). Unfortunately, the same problems that occur in backtesting, can trick you as to how well Renkos really do clarify things in live trading. When you load a chart using tradestation, esignal, or ninja (not sure about multi charts), the chart is built in a way that artificially smoothes the price action that makes it very appealing to pursue automated trading. The problem is that the same chart built on live data will loose much of its clarity. e.g. many more periods of multiple reversal bars in a row. (range bars have similar calculation problems with NT). Not entirely a deal killer, but it is one of several things to keep in mind before committing too strongly to Renko.
2. Backtesting Hell. What Mike and others have described is exactly right. NT and TS assume that the open of any given bar is the same as the close of the previous bar. Obviously this can be false in any chart type given the right circumstances, but in Renko it is ALWAYS false for reversal bars. The effect of this error in logic built into NT backtesting will guarantee not just false results, but falsely positive results. Also as stated before, the price movement outside the renko bars is eliminated by NT. This is a less serious problem than the open=close, but one you have to plan for. It is particularly dangerous in that it will let trades hit profit targets without triggering stops that would have been hit in live trading. **Both of these problems are probably solved by some of the third party renkos. I used one from purelogik that solved this.
That said, I suspect Renko could be uniquely effective trading live. If I were to pursue renko with automated trading, I would try to make sure I trade on a scale (time frame) that would keep the strategy triggers operating outside most of the noise levels. After all, filtering is what Renko charts are designed to do.
I would also like to mention that I was convinced the logic of renko charts was exactly what suited my idea of trading and I got pretty discouraged with platforms (and therefore trading) when I discovered the problems with Renko. That frustration cost me way too much time and emotional setback. I encourage you to continue to experiment with renko using market replay and enhanced third party renkos, but to keep an open mind about using other charts to accomplish whatever trading style you want to pursue and try to avoid the frustration I went through. I think you will find that the chart type won't matter much once you start 'seeing' the market in a way that makes sense to you.
If you are new to trading, you might also be focused on high percentage entries. As is discussed much here in Big Mike's, risk and trade management are critical in a trading strategy. I mention that because I think my hyper-focused Renko pursuit tended to keep me from seeing that sooner. Again, I don't want to discourage your ambitions to trade by piling on with the Renko warnings.
I have seen some of the recent posts about backtesting with Renko. This was the reason BetterRenko was created, as the focus of the bar is to have the good parts of generic Renko (visual & noise reduction), but still show accurate bar data for backtesting. Most other forms of Renko all fail back testing because they do not expose the real data to the back test engine, so the back test engine can only use what it has. By real data, I mean the actual bar open, and the wick data. Dealing with gaps/zero volume bars is also an issue (I think Ninja no longer correctly handles zero volume bars). I no longer use NT regularily (I do still write some code for others), so I am not sure what Ninja may have "fixed" recently if it is no longer working. Other platforms have similar issues when trying to deal with these bar types, so it is not a Ninja only issue. I have no intention of debugging the Ninja engine, but if there is a specific issue with BetterRenko due to something that has changed in Ninja since they were released, let me know and I will take a look if I have time.
Thanks for your contribution with BetterRenko. In my opinion, your Renko bar type works fine. I think the main discussion is to help people understand that the NT Renko (Stock Type) doesn't really reflect price action and has been a trap for many strategy programmers (as well as probably a few discretionary traders). Thanks for your offer, we will keep an eye out as NT is updated to see if they break the BetterRenko type.
I know there are issues around 0 vol bars of any type, but it is best to avoid those situations if you can (ignore any VOL=0,make the bar a bit bigger and/or avoid the low volume time frame is what I have done in the past)
In the download section on futures.io (formerly BMT), JamTheTrader uploaded a new strategy type where instead of Bar Open, Ninja will use Bar Close for its open order entry price.
I was intrigued by this. Do you think it is this simple in order to get accurate Renko backtest results?
. I've always wondered how reliable their MRP data really is. Seems like a better idea to save it with one's broker feed. Have you used their data okay? It'd be nice if Ninja would just convert saved historical data into MRP... no more "data" from those coders at NT.
.