Pivoty Slope

[Lifesys]

Hi Steve
I watch your amazing Slope creations with interest, but not sure I understand the philosophy here. Seems like catching a fun bus to no-where in particular.
If only a SL was used to test, sooner or later it will get hit - for what TP ?
As mentioned on another thread we are having great success with an earlier version (no open/shut) of SJ, 2nd time account rose >20% in 3 days, but what I did not mention is that MPTM always adds a hard Atr based TP & SL and Shirt protection is always on. "Shirt" was hit twice by retracement on Monday before 5 basket targets were hit, many times the losses. To reduce retracement bounce, when |Htf| average <0.4 (currently 0.53), turn it off.
With these Slope derivatives, if there is no planned exit, what is it aiming for?
At present 7 pairs |Htf| > 0.8, 13 pairs |Ltf| > 0.6 but DP crossovers are missing out.
"Begin with the end in mind" -1st Habit - Steven Covey.

[Xeroforex]

"Begin with the end in mind" -1st Habit - Steven Covey.

Tell that to Alexander Fleming

I think that serendipity plays a part in a lot of useful discoveries. So I'm going to play with Steve's new toy, if this is getting us in to a lot of profitable trades, then that's certainly a good start (better than a great percentage of other EA's, manual systems or signals). Then we just need to find the good exit formula.

Cheers...Ted

[Lifesys]

Thanks Ted. Serendipity traders are called gamblers!
Windfalls may come, but downpours will be lost. 'Shelley' was serendipitous in one sense but can be highly profitable used discreetly. Dozens of indicators or luck can get you in a trend, getting out ahead at some point, with minimised losses is where profit lies.

[SteveHopwood]

My demo is set up - any chart whose symbol I recognise, which does leave a few out.

Htf Slope set to 0.4/-0.4. Ltf Slope disabled. PivotCrossTimeFrame = 15. EnforceTripleAlignment = true; EnforcePivotCross = false. Everything else as defaults.

Try different settings guys and see if this cat purrrhs.

[SteveHopwood]

PS using my settings took 9 trades as soon as the markets opened tonight. Let the fun begin.....

[Jonathan2FI]

Thanks Steve,
great EA to test using the fundamentals of TMA - which I am one of the converted.
Just done some backtesting with cleaned Axi Jan-July 2012 on EJ at 2.0 spread without Piv X or Piv allignment with D 0.8 & H4 0.5 using candlestick SL on Daily -result +30% net profit with 10% RDD 6 months.
On Btesting I am looking for >50 trades, RDD <10% NProfit>15% pa & NProfit>>RDD say 2x and smooth equity curve.

This EA has legs although it only took a limited number of unique trades <10 over 6 months so can not say with hand on heart it is statistically significant.
I will test over longer period and reoptimise to see if I can hold the 30% and test 10 other pairs and then I will update all with test results when I have checked.
Thinking about possible devels (very early to be talking about this but heyho).....to improve efficiency of entries e.g. retracements and sticking with TMA as prime indicator is there any milage looking at H4 TMA compared to D1 TMA to pick up retracements? Any views?

JW
ps I still use the Pivots for target setting and SLs manual post entry...but obviously not in the auto Backtesting.

[Queenslander]

Hi Steve, I refer to your post Saturday Jul 21, 2012, 5:56PM I am still trying to understand 10.2 TMA slope v 1.4B 4:30 mq4 (zznbrm) which I gather is a crucial part of NB's system, as I see it's prominent on this post. but I didn't see in the download link on that post. I do have this indicator in my Custom Indicator list (Alpari) but it won't open on to my Empty4 Chart. I have the 10.2 TMA True Slope Nt v 7.13 (looks like a histogram) on my chart which is great. As for the one I can't open onto the chart (zznbrm) , does it have the appearance of a 2 line TMI? Hope this makes sense. I came accross from FF to your website as I had been following NB for some time, and I thank you for accepting my presence here. Brian

[garyfritz]

Random musings while watching the charts on a Sunday night, somewhat under the alfluence...

DP>WP>MP seems like a good trend filter in general. I note, though, that WP takes a giant jump on Sunday as it figures the WP for the new week. Even in a good trend, the WP often jumps within spitting distance of the DP, meaning that a slight pullback in DP could cross the WP and mess up the Pivoty trend signal. But if the trend continues, the market and the DP soon leave the WP far behind.

In other words, the DP>WP trend signal has unequal effect during the week. In a trending market, it's harder to maintain the DP>WP signal on Sunday/Monday than it is on Friday -- simply because the WP doesn't change all week. Seems to me that you would want equal trend measurement at ALL times, Monday or Friday.

And of course the same thing is true for the DP through the day, and for the MP through the month.

Here for example is a standard DP/WP indicator, using the (H+L+C)/3 of the D1 or W1 candle. Notice how the WP almost catches the DP on Monday July 2 or July 16. On Monday it looks like a weak trend (DP very close to WP) but on Friday it looks strong (DP far from WP).

Pivot1.gif

It seems to me that you'd want the same validity of trend measure regardless of day of the week, or hour of the day. That means you'd want the "WP" to update on a daily basis, or even on every candle, so it's valid for that day's DP. Make sense?

One way you could address this is by calculating the WP based on the last 5 D1 candles, instead of the last W1 candle. (Or 6 candles if you have 6 D1 candles per week.)

Or you could do like I did here: update the "pivot" calculations on every candle. The "DP" is the (H+L+C)/3 of the last 6 H4 candles (24 hours' worth), and the "WP" is the (H+L+C)/3 of the last 5 * 6 H4 candles (5 days' worth). That way the "pivot" relationships are equally valid on every candle, instead of changing throughout the day or week.

Pivot2.gif

This is a more ""accurate"" way to represent the "pivot." Is it more useful? I'm not sure.

I think it's probably a better trend measure. In particular I think it would be a better trend measure for Pivoty Slope. But I suspect it would also work better for manual NBob10.2A trading.

The normal DP is a lagging indicator, since it uses yesterday's H/L/C AND it doesn't change for 24 hours. That means we have a steady price level to use as a pullback target, which is useful. This constant-update version doesn't do that. In fact the "new DP" follows the price so closely that you really can't use it as an entry level. The "new WP" might work better for a pullback / re-entry level, e.g. on 7/12 - 7/13.

Bottom line: this may be an "interesting, but no thanks" idea. The simple flat-line DP/WP/MP calculation is easy to see, easy to understand, easy to trade with, and may be "good enough" even though it isn't really accurate. But the constant-update "pivot" calculations might provide us with quicker and more accurate trend signals.

Thoughts?

Unfortunately it's easy/quick for me to knock out these example indicators in Tradestation, and hard/slow to do it in Empty4, so I can't give you something to play with right now. If there is sufficient interest, I'll put one together for Empty4.

[monkeh]

Bottom line: this may be an "interesting, but no thanks" idea. The simple flat-line DP/WP/MP calculation is easy to see, easy to understand, easy to trade with, and may be "good enough" even though it isn't really accurate. But the constant-update "pivot" calculations might provide us with quicker and more accurate trend signals.

contrary, its "interesting, please go on" idea!
and its so simple and logical, that i start to wonder why the hell i did not think about it myself?
i feel so small with all PowerMinds up here in this forum..

thank you, Gary,

i would definitely want to see this indi for crap 4 if you ever have a spare minute!

[grimweasel]

Interesting Gary - but the whole idea of the pivot is that it shows the central fulcrum around which price rotated the previous day. It's the middle of the previous day's action as far as I'm aware? Surely, by having one that updates all the time we risk losing the pivot's true signal and merely following an MA?

I get your point however, regarding the daily and weekly jumps. But then we may as well just follow the 9 and 20 emas? I know of many institutional traders that keep it simple and just use the 9ema as a dynamic trendline and the 20ema as a dynamic pivot. So, if price is trending up, pulls back and breaks the 9ema on good volume, then we can expect a reversal after a test (usually a higher high or lower high test of the pullback high). Once this happens we are looking to go short, and we go short on price action around the 20ema.

I'm reading an excellent scalping book at the moment and all the guy uses is candles and the 20ema as his trend indicator.

We could make things simple and just show the weekly pivot. Trade long if price is above (using the daily for tighter entries on pullbacks into the weekly pivot and vice versa for short) and short if price is below. Then use our daily trend filters (9 &20emas) to trade in the direction of the weekly. Simple!