Maybe it needs to be normalized. Normalizer.mq4 or something similar.

ffsss wrote: Tue Feb 14, 2023 1:28 am Seems interesting, but the bsp indicator doesnt look that goon on my renko chart. Could you share a screenshot?

Here is what the BSP indicator looks like on a US30 chart with 5-point Renko boxes (top subwindow). The timespan of this picture is 10-11am EST today. This is kvak's version that stays loyal to the original formula. EMA for both moving averages, a moving length of 100 for both averages, and that's it.

Notice how it keeps us in the long trending moves and then gives us a better idea of trend exhaustion. Most other indicators used on a Renko chart would force you to exit a lot earlier and miss out on even greater trading gains.

Centaur at the No Nonsense Forex thread did something really interesting with the BSP indicator, which you can read from posts #153-159.

He came up with four different variations of the BSP indicator that each have a different mathematical formula, and tested them on five separate timeframes (M5, M15, H1, H4, Daily).

Centaur wrote: Tue Feb 14, 2023 12:00 am Calc1: version by Tradernawab:
Buy Pressure = EMA(close[0] - low[1], n)
Sell Pressure = EMA(high[1] - close[0], n)

Calc2: version by fract:
Buy Pressure = EMA(close[0] - min(low[0], close[1]), n)
Sell Pressure = EMA(max(high[0], close[1]) - close[0], n)

Calc3: Jimmy's Thread:
Buy Pressure = EMA(close[0] - min(low[0], close[1]) - (min(open[0], close[0]) - low[0]), n)
Sell Pressure = EMA(max(high[0], close[1]) - close[0] - (high[0] - max(open[0], close[0])), n)

Calc4: Centaur's proposal:
Buy Pressure = EMA(close[0] - ((high[1]+low[1])/2), n)
Sell Pressure = EMA(((high[1]+low[1])/2) - close[0], n)

Here are the results for Calc1, Calc2, Calc3, and Calc4.

Centaur's conclusion? "From a readability point of view I prefer calc 1 and 4, really can't see a major difference in the performance of all 4 calculations."

Here's how I would follow up with this test:

• Re-post these pictures except have all four indicators (Calc #1-4) on a single timeframe so we can do direct side-by-side comparisons. Rinse and repeat for the other four timeframes
• Perhaps do an additional M1 timeframe for the scalpers on this forum
• Clarify which BSP indicator on Jimmy's thread he used to derive the formula for Calc3. This is important because Fract's BSP indicator uses the original formula in the MetaTrader4 V1 and V2.01 versions (posted by mrtools), while the MetaTrader V2 version uses a modified formula that deviates from the original.

Great stuff happening all around! Eager to hear what other people have to say or think about this experiment.

I re-read the official page for Fract's BSP indicator and I came across an interesting discussion on how Fract would recommend using it.

And here is a summary of what he had to say:

"It's like averaging the ATR: There is no point in exceeding the length above 60. I use a length of 24 in my strategy on the 1-hour timeframe"

In his instance, the length of 24 is for both moving averages and he is using EMAs.

If you are using this indicator on TradingView and you access the indicator's settings, you'll notice you have the option to display the indicator on a higher or lower timeframe relative to the current timeframe you are viewing:


What I imagine this means is that even when you are using any BSP in MTF mode, the recommended setting is nothing higher regardless of the higher timeframe you are using.

But as you guys saw a few posts ago, the rules change when it comes to Renko charts and so far I'm really liking a length of 100 for both averages and EMA for both moving average types. If I find better settings then you'll be the first to know!

TransparentTrader wrote: Wed Feb 15, 2023 5:55 am Centaur at the No Nonsense Forex thread did something really interesting with the BSP indicator, which you can read from posts #153-159.

He came up with four different variations of the BSP indicator that each have a different mathematical formula, and tested them on five separate timeframes (M5, M15, H1, H4, Daily).





Here are the results for Calc1, Calc2, Calc3, and Calc4.

Centaur's conclusion? "From a readability point of view I prefer calc 1 and 4, really can't see a major difference in the performance of all 4 calculations."

Here's how I would follow up with this test:

• Re-post these pictures except have all four indicators (Calc #1-4) on a single timeframe so we can do direct side-by-side comparisons. Rinse and repeat for the other four timeframes
• Perhaps do an additional M1 timeframe for the scalpers on this forum
• Clarify which BSP indicator on Jimmy's thread he used to derive the formula for Calc3. This is important because Fract's BSP indicator uses the original formula in the MetaTrader4 V1 and V2.01 versions (posted by mrtools), while the MetaTrader V2 version uses a modified formula that deviates from the original.

Great stuff happening all around! Eager to hear what other people have to say or think about this experiment.

Really like the group work to optimize this indicator.

I'll get to work on the comparisons as suggested.

Centaur wrote: Wed Feb 15, 2023 7:02 pm Really like the group work to optimize this indicator.

I'll get to work on the comparisons as suggested.

On closer inspection Calc2 seems to prevent a reversal on a major pullback on all timeframes.

Centaur wrote: Wed Feb 15, 2023 7:35 pm On closer inspection Calc2 seems to prevent a reversal on a major pullback on all timeframes.

So from an EA point of view my vote will go with the version by fract, his original calculation, Calc2 in the summary, comments?

It is evident that the changes introduced by Calc3 is an attempt to get you into the trade much earlier, but I believe the advantage of Calc2 out ways this attempt.

what a great Indi indeed ,lately I am very active in one thread on ff site,1 min counter trend trade, but soon I will open journal here and explain what I am looking at and make lvls prediction ...Last 4 weeks was very good, still cent acc but I am very happy and slowly finding my edge...Generally using extremes ,watch rsi Indi and bsp Indi can be handy as well, watching 1min factually watching them all,entering on 5 min tf,distance between price and ema 9 ,and also 4h tf(on ss aj I made prediction for lvl and indeed most of the time price may respect those lvls ,general idea is that price even in strong trend in one moment will hit ema 9...I hope I will open journal here soon....And sorry guys to write in this thread,if this pictures bother you I will delete post, but maybe this can be helper for someone... .

I have to many ss ,again, will try to open new journal here,if this bother you guys I will delete these post here...Thank you guys for this great Indi again

Centaur wrote: Wed Feb 15, 2023 7:48 pm So from an EA point of view my vote will go with the version by fract, his original calculation, Calc2 in the summary, comments?

It is evident that the changes introduced by Calc3 is an attempt to get you into the trade much earlier, but I believe the advantage of Calc2 out ways this attempt.

I need to take a second look, and would love to hear the opinions of other members, but so far Calc2 seems to be the best one out of all four while Calc3 does indeed get you the earlier entries.

There is still one thing that needs to be clarified:

"Clarify which BSP indicator on Jimmy's thread Centaur used to derive the formula for Calc3. This is important because Fract's BSP indicator uses the original formula in the MetaTrader4 V1 and V2.01 versions (posted by mrtools), while the MetaTrader V2 version uses a modified formula that deviates from the original."

If Jimmy or kvak or mrtools could weigh in on this important question, it would be of great help in figuring out once and for all which BSP formula is best to use going forward.