Multiple RSIs and sharp pointy objects

[WilliamB]

The system I would like to share with you was developed from various strategies I have learned about since I began trading. I'm more of a visual learner and I like to see a setup slowly developing on my charts. The essence of this strategy is simply a moving average crossover but with a few more indicators with lots of pretty colors to help me see what the pair is doing.

Number of charts used: 2, one chart for the entry and one chart that is 4 times longer to determine the trend. So for entries off the M15 chart I would check the hourly chart and for entries off the hourly chart I would check the 4 hour chart.

Pairs: ANY
Time Frames: Any time frame from 15 minutes up to the daily for the entry. The trend time frame is always 4 times longer than the chart used for your entry. I have had tested this successfully with m5 entries but at this time frame things happen too quickly for my personality.

Indicators:
Layered RSI 3-14 with levels at 80, 70, 50, 30, 20 or 8 period CCI with levels at 200, 100, 0, -100, -200
EMA 13, 65, 200
PARABOLIC SAR
MACD 12,26,9
For the entry charts I use the RSI 3-14 layered on top of each other. In other words I use a 3, then a 4, 5 and so on up to 14 all in the same spot on the chart. If you are unable to duplicate this then an 8 period CCI works just as well. I use multiple RSIs because when all periods of the RSI bunch up and form a tight rope it usually means price will make a nice bounce. I also like to see where the 14 period RSI is at the point of entry. This will be clearly obvious when you see my charts.
For the longer time frame trend chart I keep everything the same but leave out the RSI or CCI. Using the MACD along with the SAR may seem a little redundant but i have been very successful with this set up so I will continue to do so.

RULES FOR ENTRY

LONGER TREND CHART: First I check the long term chart to determine the trend. If the MACD is above the zero line and histogram is getting taller and the Parabolic SAR is forming below the price I will only look to go long, and do just the opposite to go short.

ENTRY CHART:
For long entries the 65 EMA should be angling upward and above the 200 EMA and price must be trading above 65 EMA. MACD must be pointing up, SAR must be forming below price. The best time to enter is right after the 13 EMA crosses the 65 EMA. When these conditions are met I wait for the price to come down to touch the 13 EMA. If the candle closes above the 13 EMA, I check to see the multiple RSI are forming a nice tight rope and approching the 30 level. If using the CCI it should be just touching or getting close to the -100 level. Remember trading is an art, not a science. When these conditions are met I enter on the very next candle . If the price blows right through the 13 EMA, I wait to see if it touches or closes just above the 65 EMA. If price touches or stops just above the 65 EMA, I enter on the very next candle. I always wait for the candle close less than 10 pips from either the 13 or the 65 EMA before I enter. I do just the opposite for short entries.
I usually do not take trades when price is trading between the 65 and 200 EMA. Price action usually becomes extremely choppy so it is best to avoid it. I will also not enter the if price dips low enough to touch the SAR. Now we all know that trends do not go on forever and after the price has price has touched or come near the 13 EMA 3 times, taking any more signals could be very dicey.

RULES FOR EXIT: Since I usually risk about 10 percent of my capital, I exit after 20 pips if risking 10. If you are looking for more, then you may stay in the trade until the MACD histogram begins to fall or get shorter if going long, or when price touches the SAR.

STOP LOSS: For the USDCHF, USDJPY, AUDUSD, NZDUSD I never use more than a 10 pip stop loss for time frames up to 1 hour. I have settled on this number after reviewing over a hundred trades and finding the price only moved an average of 5.1 pips against the direction of my entry. For the other Yen crosses I would use no more than 15 pips. Now having said that, May 6 and 7 were extraordinary days and using stops this tight would have got you stopped out. So adjust when necessary.

No strategy will ever be perfect but being able to keep my stops at less than 15 pips for most pairs on the shorter times allows me to increase my leverage and grow my account much faster than before. My longest win streak has been 17 trades in a row and I have been using this steadily for the past 5 months.

The following charts from 05/07/10 is a trade I usually don't take but the signal looked strong so I pulled the trigger anyway.

-WilliamB-

[WilliamB]

Thank you for the kind comments. I've had several months where my success rate was between 80-90%. Certainly widening your stops may increase your win rate but that does not always translate into higher profitability. A good example is whats going on over at Zulutrade.com. Here you will find some of the top traders with a winning percentage as 95%. But nearly all in the in the top 5 are risking between 500 - 1000 pips. And none currently are averaging more than 12 pips. The demo account I opened with zulutrade has lost 23% and is currently down over a thousand pips after 15 days. In case your not familiar with them they're a mirroring service that allows account holders to link their live or demo accounts to the signal providers to place trades automatically via a software bridge. The top traders have hundreds of disciples so these people either have large accounts or are extremely lucky. With just my small 2:1 reward/risk I could theoretically lose the first 10 out of 20 trades, then win the next 10 and still come out with a 115% gain while still risking 10% on every trade. I've yet to lose 10 trades in a row since I've begun trading so I shall continue to use high leverage and tight stops. Now if someone is able to risk a 1000 pips and not lose more than 10% percent of their equity then obviously they have the luxury of having a huge account to take such a big hit. But it still mystifies me why people still trade that way to earn such a small reward.

-WilliamB-

[WilliamB]

Placing 2 trades a day would be an excellent strategy as I usually don't see more than 2 signals on the m15 charts from the open of the New York session until about an hour after the close of the London. I 've done some crazy stuff with my demo account such as the ones you can see on this snap shot of my demo statement. But this is breaking most of my rules as far as entries are concerned. This was a 17% gain after 4 days of trading risking no more than 10% of my capital. I wish I could have this courage when trading my live account!

-WilliamB-

[WilliamB]

That's probably a good idea. I'm sure Edward has strategies he'll never share.

[saleh1980]

Thank you william for Ur topic. It's unique and so useful.
please tell me how can I have the multiple RSI indicator or layered RSI indicator that you shown.
Thank you again.

[WilliamB]

I was able to create these with FXCM's Trading Station but I'm sure anyone with experience writing custom indicators could do it for MT4. The code for the RSI was changed when the platform was last updated but I still have copies of the old indicators. Here is a copy of the old code:

-- The indicator corresponds to the Relative Strength Index indicator in MetaTrader.
-- The formula is described in the Kaufman "Trading Systems and Methods" chapter 6 "Momentum and Oscillators" (page 133-134)

-- Indicator profile initialization routine
-- Defines indicator profile properties and indicator parameters
function Init()
indicator:name(resources:get("name"));
indicator:description(resources:get("description"));
indicator:requiredSource(core.Tick);
indicator:type(core.Oscillator);

indicator.parameters:addInteger("N", resources:get("param_N_name"), resources:get("param_N_description"), 14, 2, 1000);
indicator.parameters:addColor("clrRSI", resources:get("param_clrRSI_name"), resources:get("param_clrRSI_description"), core.rgb(255, 0, 0));
end

-- Indicator instance initialization routine
-- Processes indicator parameters and creates output streams
-- Parameters block
local n;

local first;
local source = nil;
local pos = nil;
local neg = nil;

-- Streams block
local RSI = nil;

-- Routine
function Prepare()
n = instance.parameters.N;
source = instance.source;
first = source:first() + n;

local name = profile:id() .. "(" .. source:name() .. ", " .. n .. ")";
instance:name(name);

pos = instance:addInternalStream(0, 0);
neg = instance:addInternalStream(0, 0);

RSI = instance:addStream("RSI", core.Line, name, "RSI", instance.parameters.clrRSI, first)
RSI:addLevel(0);
RSI:addLevel(30);
RSI:addLevel(50);
RSI:addLevel(70);
RSI:addLevel(100);
end

-- Indicator calculation routine
function Update(period)
if period >= first then
local i = 0;
local sump = 0;
local sumn = 0;
local positive = 0;
local negative = 0;
local diff = 0;
if (period == first) then
for i = period - n + 1, period do
diff = source[i] - source[i - 1];
if (diff >= 0) then
sump = sump + diff;
else
sumn = sumn - diff;
end
end
positive = sump / n;
negative = sumn / n;
else
diff = source[period] - source[period - 1];
if (diff > 0) then
sump = diff;
else
sumn = -diff;
end
positive = (pos[period - 1] * (n - 1) + sump) / n;
negative = (neg[period - 1] * (n - 1) + sumn) / n;
end
pos[period] = positive;
neg[period] = negative;
if (negative == 0) then
RSI[period] = 0;
else
RSI[period] = 100 - (100 / (1 + positive / negative));
end
end
end

[gavind]

is anyone else actively using mt5 already. Can you please post some pointers?

[vuanh]

yeah. i combine some indicator with rsi. we will build effectively systerm. i think each trader will has different systerm. although they have same indicator. systerm base on the knowledge and experience of trader