ADAPTING NB10.2A FOR LOWER TF & RM TRADING - 10.2A CTS

[idempotent]

Is there not enough here to go on with, without CC?

(I am subscribed, but have only skimmed so far. It's on my list of threads to look at, but I am trying to go through each one thoroughly and in turn - that's my excuse)

Who has traded / is trading CTS? How did you find it?

[KevinT]

CC took off when his last post on another thread was left hanging (by Steve)

[idempotent]

I can't see that. Might have been deleted.

Be that as it may, either the strategy had legs or not. Perhaps there's enough information from CC in the thread already for us to run with it and see if it's the real deal or not.

It certainly seemed promising. Oh for more hours in the day, and more brain cells.

[mjws00]

I'm not sure this ever really got distilled for the masses. It is very similar to how I trade, and have backtested and forward tested with pretty great results. The primary difference is I also take the countertrend retracements. I didn't find all the pivot based trend filters to be that useful when there are often 200-300 pip retracements that are linear and easy to catch. With very good RR ratios available. My testing showed they are certainly worth the extra effort and potential risk.

If you simplify the pivot ideas, you end up with something very close to extreme tma using the .8 slope value as a trigger. So where does it fall apart....?

In my testing it does exceptional during 2012 and 2013, ignoring pivots. (80% hit rate, Profit factor 2.2, 15% drawdown) 2009-2011 you can easily get stuck selling an upper, or lower band that is going to run a very long ways against you. The pivot/trend filter helps mitigate that somewhat, but you miss out on almost all of the entries that would get you back in to the trend. As often the lower band is missed significantly.

It is not really any more remarkable than any other band based system. This is classic reversion to mean trading with a pivot based trend filter. But in the last 18 months or so many forex pairs seem to respond very predictably to atr based bands. Nice sharp turns with little draw down. CTS went one step further by reducing risk to only taking trades in its defined trend direction. The result is fewer trades, a higher hit rate, but overall (at least in my manual and ea based testing) reduced returns. This could be mitigated with leverage, but you have smaller sample sets and will be burned much harder when drawdown hits.

I have live traded something similar since last sept. Hit rate is 80%ish. With most trades being profitable if a Jump stop used. With std stops I could get down to 60 pips trading manual, needed 150 to take all trades mechanically optimized (ea). RR drops, but hit rate is so strong it is still very profitable.

Was sorry to see the thread die. Baluda's integration with MultiBob would have been fantastic. I'll throw up my quick and dirty ea for backtesting and playing if I get a chance.

[mjws00]

Here was the quick and dirty ea I built for testing the entries from this and al's extreme tma. Remember these are longer swings 2-3 days, so tick modelling etc is not a problem.

Super simple... sell outside the 240 band. Every time 1H slope is less or greater than +-.8. Exit at cross of opposite band. I set parameters for a delay between trades, so if it was running against us at the top of the bands we would wait before selling again, and max number of open trades.

I played with a whole bunch of parameters and ranges and it really requires very little to get very good returns when the market is behaving. 18 months 100+ trades. There were a bunch of longer periods in 2008-2009, where the bands would get pegged and not trigger a positive exit as coded. A jump stop would fix a lot of these as they tend to go profitable in an extremely high percentage of cases.

Testing by hand, I never got caught in these draw downs, but the simplified robot logic does. Chances are coding in the additional CTS logic would save most of these as they are always countertrend trades.

I am using 240 tma, 1H back to +-.8 as a primary setup live since Sept. It is great as an alert for a potential top or bottom. But it is always EARLY. So you need a filter or trigger to slow it down, or position sizing, or multiple entries to average in to get the best price. HA candles, MA cross... pick your poisons. I add Supply and Demand, ATR, and Pivots with discretion, but often find a simple MA cross I like 7/21 on the 30 min... keeps you out of trouble. These are 1-3 day trades often for a few hundred pips. RR can easily be 3:1 or better, and you will be in with a great entry when they run.

You will swing band to band (use wiggle if you want) in a very very high percentage of setups. The gotcha is you will often have one or more swings of drawdown before the big move. If you are used to multiple entries cool. If you have proper position sizing and money management... cool. If you overleverage and run tight stops or panic... you will lose your shirt, and watch in frustration as it hits your take profit a day after you punched out. It is a good candidate for a limited recovery type system ala lorente's extreme tma. I prefer multiple BE trades with re-entry vs the big drawdowns. I take half off at SR1 and go to BE. I always allow for dual entry, I consider a 3rd swing an error and don't let positions get heavy. I tested potential 4 level recovery, it works great with small positions and in theory... at full weight I personally can't stomach the drawdown.

Experiment. There is a reason this style of trading has had such longevity. It is well researched. The only things changed are ATR Bands vs say Bollinger or Channels, and NB's trend definition by pivot.

I've had a lot of fun with this one and continue to trade it.

Mike

P.S. DO NOT TRADE THIS EA LIVE. IT IS A PROOF OF CONCEPT HACK!!!!

[idempotent]

I'm sure there was some english in there somewhere

mjws00, that's a lot of info. I wonder if you want to start your own thread, and take some of us through it.

Thanks for the posts so far