Win ratio

Here is my latest chart update on the Cable (GBP/USD) >>>

As before…

Explanation

The chart is from GFT Dealbook and processed in Microsoft Paint.

I switched the time scale to GMT+1 (easy).

The pink areas are non trading time.

The white areas are trading time, in this case 5.00 am to 1900 pm GMT time. (London open).

The green lines extend from the highest non trading pink time into the white trading time. These lines mark where to trade long in the white trading time.

The red lines extend from the lowest non trading pink time into the white trading time. These lines mark where to trade short in the white trading time.

If the price action did not trigger the trade in the white trading time, a red cross is marked.

The other trades were triggered and in each case drew large amounts of pips.

This means that my suggestion of using the multiple lots positions approach would be very successful with this system using an initial 30 pip risk, 10 pip first target and then 30+ for second target.

You would have to monitor your trade, but your profits would be very much better.

Trevpick wrote:

“if the highs or low are are near or are at the beggining of the time zone then use that as the high” (copy/pasted)

my mistake, my interpretation of the above statement was wrong.
I taught since the first candle was the lowest low, it would be also my lowest one to draw a line 6 pips below. Vice versa, if it would be the highest one I use it as the highest 6 pips above.

In you picture you are a winner. Glad I slept fine. :smiley:
Happy Trading

Lol - how many different methods on this post!!!
I think you’ve got your work cut out :slight_smile:
Do you backtest on GFT? I need to learn how to do this.

Methods:
Trail by 5
Trail by 10
Target all at 10% daily ATR
Close half at 10, close half at 30+
30:30
I’m trying: close half at 10% ATR (about 26pips), then trail half by 10% ATR amount with a possible manual close. SL is 1.5x 10% ATR. Less than 1:1 to start with but the half lots closing might allow more.

Another successful trade this morning.

I agree, amazing in how many ways infos (instructions) can be understood by different people.
Since Oanda has a minimum of 10 pip TS, it solves the problem with a tight TS. Meaning, either you set the 10, do manual trading or open a account with Alpari. :smiley:

If you learn how to backtest on GFT, you can teach me!! :o :wink: :cool:

I have not the faintest idea on how to backtest!! :o

I’m sorry to have to ask but I want to be absolutely sure I’ve got it right:

Trev, is your overnight period (on 1min scale) from the 1900 candle to the 0459 candle (British summer time) (European Central Summer Time (GMT+2) that corresponds to 2000-0559

Reason I’m asking is that I became unsure whether it was 1900-0459 or if it was 1900-0559? (BST)

Tymen,
I was thinking about GFT but most brokers use MT4 and most stuff (EA) is written for MT4 in all the forums. (Except your weather system, nobody converted it to MT4. brrrrrrrrrrrrrrrrrrrrr) :smiley:
It’s just like PC versus Mac, maybe Mac is better but PC is the winner for now, so I go with the flow. :slight_smile:
I actually really liked Ninja (except the $900) but hardly any brokers are out there to make use of the full potential of it.
Happy Trading (Gambling?)

GMT 18:00:00 to GMT 04:00:00.
I followed Phil and have demo IBFX (GMT time), FXDD demo but it is not GMT so I let it dye, Oanda is for trading, set to GMT, I hate there chart-system something a kid made in his spare time. :slight_smile:


draw two lines, one at 19:00 London time, and 1 at 05:00 London time

I in fact drew my vertical lines on the 15M chart at 1845hrs & 0515hrs so
I could see the 1900 & 0500 candles.

But as with all things Forex nothing should be taken as black
& white, there are so many grey areas, try to demo all of the
various permutations.

(Or just wait for Phil to publish his results :))

I’ve been trading Trevor’s strategy this week, and I’m starting to really like it. I’m trading it exactly as he first outlined it,
except for the trailing stop, and a slight tweak on the buy side.

Here’s the full monty:

In my eastern U.S. time zone, Trevor’s “overnight” period is 2pm - midnight. At midnight, I check the GBP/USD 5-minute chart, identify trade-able high and/or low prices in the “overnight” period, and place my entry orders.

I place an entry order to buy, 6 pips above the highest [B]Ask[/B] price during the “overnight” period.

I place an entry order to sell, 6 pips below the lowest [B]Bid[/B] price during the “overnight” period.

My trade size is 10 times the balance in my account.

I place a 10-pip TP, and a 30-pip SL (and I do not use a TS). My TP represents 1% of my account balance. My SL represents
3% of my account balance.

Then I go about my business, and let FXCM’s server do its thing.

It will take months to prove whether Trevor’s claim of a 98% win-ratio is true. But, so far this looks like a sound strategy.
And I’m comfortable with the 3:1 upside-down risk:reward ratio. At 3:1 R:R ratio, a 75% win-ratio would represent break-even.
Anything above 75% would represent overall profit.

In my limited experience with this strategy, it seems to me that success depends very much on identifying a good, trade-able short-term high and/or low in the “overnight” period, OR staying out of the trade.

If this strategy continues to work, I’ll experiment with increasing my lot size, and increasing my TP.

I’d like to hear how you guys are doing with this strategy.

Clint

It’s easy to find out:
go back to page two in this thread, there you will find his post with a thumbnail attached, right click the thumbnail, say "open in new tab or new window. Then look very carefully over his trading that day. The chart is a 30min chart.
trevpick is counting his money right now, No time in here :smiley:
Hope this helps.

I have made some slight modifications to it as well but will test it out every day over the coming month. I am trying it on the USDCAD as well but have adjusted TP and SL smaller.
However, my modification, to adjust for bedtime :), is that I work out the highest high and highest low from the close of NYSE to the open of London. I guess if you’re being picky, it should really be the open on Frankfurt but haven’t seen too much action happening between 0600 and 0700GMT (0700 and 0800British summer time).
My 1st TP however is 10% of the daily ATR and then I trail by 10% of the ATR. I’m thinking it would be better to trail by a smaller amount or just close at 20%.
I do take account of any recent S&R lines in the trade and as long as there is room for some pips I’ll trade both sides of the breakout.

I’m interested in Phil’s opinion as to why he thinks this won’t work in quieter markets. Isn’t this similar to the Big Ben strategy?

Ok, I’ve tried out a bunch of 1:1 SL and TP numbers and, if you want to go that route, go with 30 pips.

A 30 pip SL and TP would have gotten you a 59% win rate for the year with 64 losers and 94 winners. This is based on taking all trades that come along and not using any type of filter.

And before anyone says a 59% win ratio isn’t any good just remember that comes out to be 30 more winners than losers. If you traded 2% of your account per trade that would be a 60% gain this year. :slight_smile:

If you wanted to filter the trades I’d suggest skipping the trade on Mondays. I guess it’s because the bankers are late to work on Mondays or something, but these trades fail more often than on other days. I’m just guessing based on observation, not hard numbers, but I think the win rate would probably go up to 65% if you skipped Mondays.

I’m still convinced the secret to this system is in finding a way to calculate SL and TP based on current market conditions instead of a set number. In the beginning of the year when daily ranges were larger 30 pips did fantastic, but as the year went on and the ranges shrunk the number of losing trades shot up.

I’m going to test this some more using various ATR values instead of set-in-stone numbers.

And for Tymen and others that what want to know how to backtest, it’s simple but time consuming. I’m just scrolling a 5m chart back to the first of the year and inspecting the days one-by-one to see if they win or lose. :slight_smile:

Is there a way to calculate the exact R:R on trades where you close half at a 1st TP and then trail?
If I was using a 1 : 0.67 risk:reward (3:2), it would need to be successful 60% of the time, is that correct?

Given a day or 2 I could probably drum up something for the MT4 strategy tester. Speaking of which, does Dealbook have a strategy tester?
[I]Edit: Yes, Dealbook has a strategy tester but it’s written in some god awful language called CTL, which looks like something from the 1980s. Give me MTL4 code anyday :)[/I]

Base your calculations of Risk and Reward on INITIAL conditions at the time you enter your trade. Those are your estimates of
(1) maximum acceptable loss, represented by your initial stop loss, and (2) potential profit, represented by your initial take-profit point.

In other words, R:R calculations ignore changes to your SL or TP after the trade is underway. Those changes (when they occur) are referred to as trade management, and they include: manually adjusting your stop, trailing your stop, adjusting your TP, and selectively closing a portion of your position.

A R:R ratio represents your plan, made before you enter your trade. In the simplest kind of trade, your actual results might match your calculated R:R ratio. Here’s an example. You enter the same type of trade every time, including a 20-pip SL and a 30-pip TP. You never intervene in your trade. Instead, you let it run until either your SL or your TP takes you out. Clearly, in this example, all of your trades will produce either 20-pip losses, or 30-pip profits; and all of your losses and profits will exactly reflect your initial R:R ratio.

In actual trading, this is rarely the case. Trade-management actions change the actual losses and profits you end up with. It’s a good idea to analyze these actual results from time to time, but it’s a tedious task requiring some time and some math.

In order to compare your actual trading results to your intended Risk:Reward ratio, you have to determine your average loss, and your average profit, over a certain period of time. Print out a copy of your broker’s statement showing every trade you made. Isolate the time period you want to analyze — say, the previous month. Cross out all the break-even trades. Total all the losing trades, and calculate the average of these losers. Total all the profitable trades, and calculate the average of these winners. You now have two averages. Divide the larger one by the smaller one, and write the result as a ratio. You can now compare this ratio of actual results to your initial Risk:Reward ratio.

Example: you had 16 losses totalling 360 pips, and 25 winners totalling 730 pips. Your average loser was 360/16 = 22.5 pips. Your average winner was 730/25 = 29.2 pips. Dividing these two averages, 29.2/22.5 = 1.3, so your actual results were
average loser : average winner = 1 : 1.3 over the past month.

Note that it’s common practice to refer to a risk:reward ratio, not a reward:risk ratio. And I have kept things in that order in the examples above.

Correct. 60% or your trades would have to be winners, in order to break even overall, [B]in theory[/B]. But, as was shown above, your actual results may be different from your intended R:R ratio.

For the algebraically inclined, here is the math.

Let W = % winners.

Then, (W)(0.67) - (1-W)(1) = 0 (break-even)

0.67W - 1 + W = 0

1.67W = 1

So, W = 1/1.67 = 0.5988 = 60% (rounded off)

You never cease to amaze me Clint! Thank you very much for posting this!

Thanks, Phil.

I think a " Reward/Risk " ratio of at least 1: 1.5 is realistic and easy to achieve.

With a proven strategy with accuracy of 60%…

It is really good enough to see positive profits of at least 150-200Pips monthly…

Furthermore…if you guys trade the more volatile pairs such as the GBP/Usd where breakouts are very common…

You can literally trade just 2 -3 times a week and grab some serious pips too.

Bottom line is…

Using a good reward/risk ratio of at least 1: 1 is very crucial for long term success…no doubt about it.

But I personally go for at least 1.5 - 2 x the rewards…

Hope this helps…

[quote=phil838;114020][/quote]

Hi Phil.

I just wanted to ask if you’ve had the time to do any work on replacing a fixed TP/SL with numbers based on ATR?

I’m going to investigate that myself but lack of time will not allow it for some time, so I was wondering about getting a bit of a free ride with your research :smiley:

Not yet. I’ll start messing with it on Monday and let everyone know what I come up with. :slight_smile: