for the life of me, I dont see how to download these unirenko charts? can someone send a link?

I was playing around with this bar type and i noticed that when I put a strategy on a chart and look at the historical performance, the orders are not filled at the same prices as when running realtime (or market replay). In my strategy I have calculate on bar close true. In realtime, the market order is sent on the bar close price. In historical, it's sent on the bar open price. If the open setting is, for example, 3, then it can send the market order at a price that is 3 ticks better. This gives a 3 tick edge in the strategy performance but realtime it's 0.

I didn't see this mentioned in the thread, so I'm curious if others have seen this, and if there is a way to work around this? Like using a secondary bar period of 1 tick or something?

@Nowshad: Consider Renko bars as an indicator. Renko bars do not show real prices.


Some properties of Renko bars:

-> when price continues to move in the prior direction the open of a Renko bar is close to where the first tick actually plotted
-> when price reverses direction compared to the prior bar, the open of a Renko bar is far away from the first tick that plotted because the reversal is not shown on the chart
-> in most cases a Renko bar will not show the correct high
-> in most cases a Renko bar will not show the correct low
-> only the close of a Renko bar will always be near the last tick of the time period covered by that Renko bar

Implications:

-> you cannot apply any indicators to Renko bars that use open, high or low as input
-> for example an indicator that shows daily high and low values will always show false values
-> a Fibonacci indicator will always use false swing highs and swing lows
-> you cannot do a proper backtest with Renko bars, because the strategy typically enters a position at the open, and the open shown by the Renko bar is false


Displaying real open, high, low and close of a Renko bar:

If you wish to display the real open, high and low of each Renko bar, this is pretty easy. @aslan has done the work for you and coded a bar type, which is called BetterRenko. Make sure that you install the BetterBrick chart style along with the bar type. With Better Renko you will be able

-> to show the real open, high, low and close that was made during the bar period
-> to apply indicators that use open, high or low as input
-> to apply an indicator that shows correct daily high and low values
-> to apply a Fibonacci indicator which will use correct swing highs and low
-> do proper backtesting


.

I was playing around with this bar type and i noticed that when I put a strategy on a chart and look at the historical performance, the orders are not filled at the same prices as when running realtime (or market replay). In my strategy I have calculate on bar close true. In realtime, the market order is sent on the bar close price. In historical, it's sent on the bar open price. If the open setting is, for example, 3, then it can send the market order at a price that is 3 ticks better. This gives a 3 tick edge in the strategy performance but realtime it's 0.

I didn't see this mentioned in the thread, so I'm curious if others have seen this, and if there is a way to work around this? Like using a secondary bar period of 1 tick or something?

I updated my strategy to submit all market orders on a second bar series with a period of 1 tick. I'm on NT7. This fixed most of the discrepancy between historical & market replay. I compared 1 day's executions, which I admit isn't much, and I got deltas:

entry delta exit delta
0.01 0
0 0
0 0
0.01 -0.01
0.01 0
0.02 -0.01
0 0.03
0 0
0 0.04


0.05 0.05

total delta 0.1
avg delta 0.011111111

so now I can plug 1 tick into the strategy slippage parameter and get results similar to market replay.

I tried better renko and got similar results, so it seems executing market orders on the 1 tick bar series did fix the issue.

Thanks for the feedback everyone.

You will now get back test quality on 1 period range bars, faster than 1 period tick ! , as long as you have tick data in the Ninja database.

In this thread, I am posting my Universal Renko bar type implementation. I have been using it exclusively in my trading for a while now, and have had many requests for it. I wanted a more flexible Renko bar that could be customized in a particular way to fit my trading style, so I decided to make a bar type that allows the user to configure it more liberally to better suit their particular needs.

Basically the UniRenko bar, allows the user to choose when the bar closes in the direction of the previous bar (Trend), and when it closes in the direction against the previous bar (Reversal). So, you may choose for a new bar to form when the price has moved 5 ticks up from the previous close, or 10 ticks down from the previous close.
This allows you to tailor the bar in ways that hopefully may enhance your particular trading style. For instance, I am a reversal trader, entering on reversal bars, and putting my stop behind the entry bar. So, I have tested and chosen trend and reversal settings that minimize the size of reversal bars, and therefore minimize my stop loss sizes, while still producing clean sharp indicator patterns my trading method is based on.

Like most of the other advanced Renko bar types, the UniRenko bar open is artificial, so I consider it aesthetic only, and I do not use it in the bar size calculation. The bar calculations are based on the previous bar close. I personally like to place the open half way up the previous bar as I like the way the bars look with that setting. The tails on the bar show real prices. The bar should be backtestable if stops/targets are placed outside of the bar.

Because all the parameters are user selectable, and no aspect of the bar is derived from an arithmetic calculation, there should be no restriction on any bar parameter having to be odd or even, etc. Internally uses NT PeriodType.Custom5

I have hesitated to post this bar type before, as it can be configured to mimic some commercially available Renko bar types. I purchased some of those commercial bars myself, but they did not fully meet my needs. The purpose of this implementation is to provide traders with more flexibility, that hopefully with some experimentation may ultimately enhance their trading like it did mine.

futures.io (formerly BMT) Download link:

Where is the link?