Andrew's Pitchfork Trading Strategy

[nit2wynit]

I'm clearly late to the party again.  Looks like some great info being thrown down in here.

So, are you winning now?

[DSchenk]

2 minutes ago, nit2wynit said:

So, are you winning now?

The algorithm is

I haven't tried on live yet, but hope to be ready to trial this from tomorrow. will obviously update how it goes  

[nit2wynit]

1 minute ago, DSchenk said:

The algorithm is

I haven't tried on live yet, but hope to be ready to trial this from tomorrow. will obviously update how it goes  

Good Luck (but luck it for Gamblers sooooo.  i wish you all the best of success with it.

I'll read back thru it all. 

[EugeneB]

Wow. That looks crazy good. I've been looking at the coding to understand it better and I can't seem to find out where the problem lies. 

[EugeneB]

Anyone that can recommend any tips for me optimize the code?

[DSchenk]

So I was crunching the numbers on this yesterday for FTSE 100, DAX, NASDAQ, NG and CL.

Table below shows how it looks.

Definitions:

TPL (Take Profit Long)

TPS (Take Profit Short)

X (Points above L0 - Low of previous Day for entry Long)

Y (Points below L100 - High of previous Day for entry Short)

 

Considerations:

While the results do look promising, I'm having a few concerns, which I want to iron out.

1) I'd rather have all X and Ys in the positive area, so that the entry is always in between L0 and L100. That just sounds more intuitive, but the data obviously needs to tell what's better. DAX is currently set-up that way.

2) The algorithm currently increases the position size as a function of the account size. After reconsideration I think it would make more sense to calculate based on a flat size, so that all trades in the sample size of 100,000 units are weighed equal. With increasing position size, the more recent positions count way more than the positions in the beginning, which may skew the results. Obviously the % gains in the end will be less, but we know when actually trading it, it will be much higher. (Even more so if you take into account you trade multiple of these strategies the same time, this leads to a further multiplication effect of profits)

3) Currently there are no stop-losses applied in the strategy, which doesn't seem right. Looking to find a way to work stop losses in without sacrificing too much profitability

4) I took an actual live trade today on DAX with this strategy. Screenshot below. Entry at break of the red line at 7.55, which was bit of cheating as strategy only starts at 8.00. (Correct entry would've been the candle at 8.15 breaking through the red line again.) Profit target would've been about 18 points below the Pivot, unfortunately we saw a bounce off the pivot. So I'm asking myself would it not have been better to either take profits at the pivot or at my daily goal of 20 points (on the way back up). Need to see if I can work stuff like this into the strategy.

 

Ticker	TPL	TPS	X	Y	Result over 100,000 5min units	% of winning trades	Entry from	Entry until	Flat after	Time in Market
FTSE 100	34	48	-30	8	1466%	65%	08:00:00	16:00:00	21:55:00	19%
DAX	30	55	14	22	1131%	71%	08:00:00	16:00:00	21:55:00	22%
NASDAQ	85	99	-9	-37	2065%	59%	10:00:00	21:00:00	21:55:00	10%
NG	45	41	-42	6	1864%	59%	13:00:00	20:00:00	21:55:00	11%
CL	95	85	37	-15	1147%	59%	15:00:00	18:00:00	21:55:00	10%

 

 

[DSchenk]

Trying to simplify the strategy proves to be a rather complex exercise.

Not sure if on the right track here, or lost in the woods, trying to find a tree.

It should be as simple as finding out the thresholds where an index becomes overbought, then entry short. Where it becomes oversold, entry long. Take profit and stop loss a function of historic data.

Backtesting this is fascinating though . Only small deviations in parameters can lead to a huge difference in the end result. Like changing the entry level by a single point - couple of 100% gain lost.
Maybe that's why 95% lose in trading, the margin of error is astonishing. Half a point entered too late or too early, profits halfed. Half a point exited too early or too late, profits gone. Stop loss set half a point too far or too tight, huge losses.

[nit2wynit]

18 minutes ago, DSchenk said:

Trying to simplify the strategy proves to be a rather complex exercise.

Not sure if on the right track here, or lost in the woods, trying to find a tree.

It should be as simple as finding out the thresholds where an index becomes overbought, then entry short. Where it becomes oversold, entry long. Take profit and stop loss a function of historic data.

Backtesting this is fascinating though . Only small deviations in parameters can lead to a huge difference in the end result. Like changing the entry level by a single point - couple of 100% gain lost.
Maybe that's why 95% lose in trading, the margin of error is astonishing. Half a point entered too late or too early, profits halfed. Half a point exited too early or too late, profits gone. Stop loss set half a point too far or too tight, huge losses.

In too early, out too late, or in too late and out too early...... Some with larger accounts might say that it matters little so long as the Trend Direction is correct and the Stop is adequate; if they even use one.

I've said this before.  Account size limits how long you can play.  If you have millions to spare, you could just put a massive Stop and go in any direction, (if you don't have the time to wait for the perfect In Point) so long as you have belief in the Trend Direction.  What goes up must come down.  With a small account I am forced to play a short game OR I can play long but less often, but this will tie up my funds until I see a decent profit.  Will a small account I can't justify seeing a profit of 20 points for it then to come back 10 or 15, then hope it goes back up.  I would never allow myself to lose 50% of a profit if I saw it on screen.  This is why I play 1:1 PL. and hope for 1:2 or better.  The larger the time frame the larger the Stop therefore the Larger or Equal the Profit must be.  I know this is basic stuff, but behind the knowledge is our Psychological Approach to it also.  You are clearly Intelligent enough to break it all down and make use of the info.   I just look at the chart in front of me.  

As we've discussed in the past, Indicies will go up 20 points and come back 15 points and go back up 20 points and come back 15 etc, so on and so forth.  This will obviously happen more so on a 1M or 5M, but on a Daily or Weekly chart, you'll see little difference in movement, with your timeframe for Profit being Days or Weeks respectively.  But playing on a Daily or Weekly chart requires either a massive Stop to cover Intra Day oscillation OR a waiting game to make sure you find your Support Resistance points to get in.  Waiting for a Pattern to finish or Reverse means you're not trading, which is fine so long as the profits outweigh the Time and Effort or Risk.


What I'm really saying is this.  If your Success is 50 60 or 70% then this really dictates what your Time vs Risk is or should be.  If you're only Winning 50% of the time, then your PL should be 50% so you break even.   What this literally means is if your Stake or Risk is £50 then your Profit or Limit should be £50.  However, as I've said above to Risk = Stop = In point;  The less you have, the more precise you need to be with your In Point and the Shorter you need to be in the game.

With so much Up and Down on the 5M and Indicies, how can you justify the Up and Down Profit Loss?  If it goes Up 20 points and you're up £100, would you then allow it to come down to only +£50 in the hope it will keep doing this all day, or Take Profit?

 

[DSchenk]

Good points as always @nit2wynit. You understand the game.

In all the examples you give you have declared certain variables, like this quote:

Quote

Indices will go up 20 points and come back 15 points and go back up 20 points and come back 15 etc

My answer is, if historic data suggests that it is more profitable to hold through 15 points pullbacks and then ride the next wave, then that is what should be done. If the data suggests, better to take profit after a 20 points run, then get back in on the bottom of the pullback, then this is what should be done. But possibly the optimum lies at only allowing 13 points pullbacks and then stop out or taking profit already at 17 points.
Once you have simulated every possible outcome, there are no opinions left, no emotions, just simple, plain data driven trading.

Difficulty is to build a trading framework where all possible variables can be simulated (and also having a powerful enough simulator - PRT seems to come quickly at its limits, both in terms of number of possibilities which can be simulated and processing time - simulating 200-400 different possibilities takes a matter of minutes, but possibly this exercise need to be done with millions of possibilities. That's essentially what the Quant Trading Funds do.

 

That's another good quote.

35 minutes ago, nit2wynit said:

I am forced to play a short game OR I can play long but less often

This comes down to time in market, which needs to be minimised. Better to take 10 points profit after 15 minutes, than 20 points after 3 hours. Cause you could've made another 10 points 4 times over somewhere else in that time frame.

 

However, I slightly disagree with below statement.

37 minutes ago, nit2wynit said:

Account size limits how long you can play.

Account size only affects the size of your trades. All strategies should equally work across all account sizes, by just scaling the position sizes accordingly. (Obviously there's a minimum as you have minimum position sizes defined by the broker). But it shouldn't really matter if you have £1k, £10k or £100k in your account, this just means you have position sizes of let's say 5, 50 or 500.

[nit2wynit]

58 minutes ago, DSchenk said:

Account size only affects the size of your trades. All strategies should equally work across all account sizes, by just scaling the position sizes accordingly. (Obviously there's a minimum as you have minimum position sizes defined by the broker). But it shouldn't really matter if you have £1k, £10k or £100k in your account, this just means you have position sizes of let's say 5, 50 or 500.

OK, let's prove this point.  Minimum size with IG is £250.  Look at the FTSE now but take it from 5pm.  It's only moved about 12 pt But there are several moves within it that are already decent but it retraces almost the entire amount.

I'm preaching to the choir here as I know we both know what we're talking about in this instance, but it proves that all theories and tests are specific to the pattern and Trend at the time.  I think this explains the PL ratios'.  Some people are stuck on a system, but it's requires an overall understanding of chart movement and the ability to adjust strategy to suit.


FTSE after Hours 2pt Spread.  Minimum stop 8pt.  £250 account size.  The point in this example is the limits of the Spread and the Timing.  How much movement can I expect at this time of day?  Very little, or did I know that drop was coming?

 

Edited October 22, 2019 by nit2wynit

[nit2wynit]

I've just taken profit on it at £4.03.   My 8pt Spread put me at risk of £5.?? (missed the pence) but I would only have let it reverse 4 pts or about £2 or £3.  I took profit at 7215 as i figured it was going to Stop and retrace, but I'd also figured I had my 1:1 ratio already and didn't want to risk the retrace and lose profit.  this is based on the timeframe of 1M and seeing how much the chart has been moving in 1M and 5M.

So My Potential Risk was £5.50 ish.  My actual Risk was only £2.50 ish, therefore my profit was to be £2.50 or above.  I took £4.

Edit:  Looks like I got out right time.  However, it's in the middle of the SR zone now so I wouldn't go back in till it hit one or the other.

Edited October 22, 2019 by nit2wynit

[DSchenk]

Makes sense what you say, @nit2wynit, but where's the correlation to the account size?

With a bigger account you could've risked £55 and made £40 and with an even bigger account you could've risked £5500 and made £4000. % stay the same, just the pound value changes. But pound value is irrelevant for any strategy. The more you put in, the more you get out, simple as that.

 

[nit2wynit]

11 minutes ago, DSchenk said:

Makes sense what you say, @nit2wynit, but where's the correlation to the account size?

With a bigger account you could've risked £55 and made £40 and with an even bigger account you could've risked £5500 and made £4000. % stay the same, just the pound value changes. But pound value is irrelevant for any strategy. The more you put in, the more you get out, simple as that.

 

It was more to do with Timeframe and when to get in and when to get out.  Compare it to the Daily.  It's all green.  Yeah you're right.  If you had 1mil to put down you could play a shorter and make massive profits.  My example was to show when to expect a retrace or a Pullback or a Pause in relation to Chart pattern and Timeframe.  There's lots to see and gauge on a 1M that some belive to be noise.  I don't however. 

In relation to my previous info, I suggested that the longer you're in, the more it can change.  That unless you know the Trend Direction and can cover a stop it doesn't really matter where you get in, so long as your stop is covered.  So in turn I only have £250.  I was only able to get in on this because i had a belief it had hit Resistance and was coming down.  But the Trend for the Day is Up.  If we look on the Daily I'd assume it was going to continue UP in a Trend, therefore I would bet UP with a larger account size.  the move down would be of little significance.

I think I've gone about this in the wrong manner lol.

This chart suggest such a small opportunity for profit based upon the chart movement at this time.  

Edited October 22, 2019 by nit2wynit

[nit2wynit]

Look at the Daily.  If I had only £250 but wanted to Trade today and make sure I was covered, I'd put a Stop at S3 and take profit at 7500, but with £250 account size the Stop would cost £87.  That's a massive amount to risk of only £250, plus the time it would take to reach 7500.  Therefore, the Risk and Stop is relative to account size.

The chart patterns are the same, but what time frame, and account size dictate how and where you place your trades.

[DSchenk]

I think I got your point now. You are confusing £ profits with % profits.

When you say something like 

Quote

If you had 1mil to put down you could play a shorter and make massive profits. 

What you mean are probably massive £ profits. Whereas in fact you would only make tiny profits, % wise.

Therefore I don't understand why you think, that with a larger account you would go long, whereas with a smaller account you would go short.

The goal is not to make £100k in profits with a £10m account, that's only a 1% profit. Far from massive.

The goal is to make £100k in profits starting with a £2k account. That's the same £ value, but here we're talking massive profits.

3 minutes ago, nit2wynit said:

Look at the Daily.  If I had only £250 but wanted to Trade today and make sure I was covered, I'd put a Stop at S3 and take profit at 7500, but with £250 account size the Stop would cost £87.  That's a massive amount to risk of only £250, plus the time it would take to reach 7500.  Therefore, the Risk and Stop is relative to account size.

Doesn't make sense to me, because if you had £250k in your account, then your stop would cost you £87k. It's the same %. If you don't want to risk £87 of £250, then you also probably don't want to risk £87k out of £250k, right? Which means your strategy doesn't fit you.
If you want to risk less, just trade less size. (I mentioned earlier, IG puts a minimum size in place, so we have to have a minimum account size. above that size won't matter anymore)

[nit2wynit]

I've just made another £7 on the FTSE Drop.  My Minimum Stop was still 8pt but i was right at the top of the resistance point.  I got out right away to take a clear profit.

I suppose what I'm really commenting on is how we all see different things on screen and how we play it out.

 

[nit2wynit]

10 minutes ago, DSchenk said:

I think I got your point now. You are confusing £ profits with % profits.

When you say something like 

What you mean are probably massive £ profits. Whereas in fact you would only make tiny profits, % wise.

Therefore I don't understand why you think, that with a larger account you would go long, whereas with a smaller account you would go short.

The goal is not to make £100k in profits with a £10m account, that's only a 1% profit. Far from massive.

The goal is to make £100k in profits starting with a £2k account. That's the same £ value, but here we're talking massive profits.

Doesn't make sense to me, because if you had £250k in your account, then your stop would cost you £87k. It's the same %. If you don't want to risk £87 of £250, then you also probably don't want to risk £87k out of £250k, right? Which means your strategy doesn't fit you.
If you want to risk less, just trade less size. (I mentioned earlier, IG puts a minimum size in place, so we have to have a minimum account size. above that size won't matter anymore)

Hold on.  lol.

Wait.............

£250 Not £250,000!

 

Points moved vs % moved doesn't matter.  The larger your account size the larger the profit.  Whether you measure it in £ or % makes no difference.  If i could make 1k in 1pt move or 1% move!  What's the difference.  they both equate to Time inthe Game.   the chart either moves large points or large %.  A £1 stock that move 100% still only moves £1.  But a £100 Stock that moves 1% is 100pts.  (or whatever the sum equals)  you get my point.

My goal is to use Capital to create more money.  Ross Camera doesn't Bet Exponentially.  He doesn't use 10 million to trade.  He still has a fixed amount regardless of his account size.




 

Edited October 22, 2019 by nit2wynit

[DSchenk]

Appreciate the discussion. Some good points being made.

3 minutes ago, nit2wynit said:

Whether you measure it in £ or % makes no difference

My last statement for tonight, but strongly disagree with this. It makes all the difference. Measuring in £ doesn't make much sense at all

 

6 minutes ago, nit2wynit said:

My goal is to use Capital to create more money.  Ross Camera doesn't Bet Exponentially.  He doesn't use 10 million to trade.  He still has a fixed amount regardless of his account size.

Ross Cameron account size is always the same. He starts each month with about $100k, hence he has the same targets for each month. If he would leave all the profits in his account, his targets would grow each month. Although the % profits would still be the same each month. % counts, not £.

[nit2wynit]

14 minutes ago, DSchenk said:

Doesn't make sense to me, because if you had £250k in your account, then your stop would cost you £87k. It's the same %. If you don't want to risk £87 of £250, then you also probably don't want to risk £87k out of £250k, right? Which means your strategy doesn't fit you.
If you want to risk less, just trade less size. (I mentioned earlier, IG puts a minimum size in place, so we have to have a minimum account size. above that size won't matter anymore)

No no no lol.

I WOULD NOT trade the FTSE with only £250 as to ensure I stayed in the Long Game I'd need to put a Stop at S3, when using the 1D chart.  We can see the chart on the Daily and either think it's actually going Up still coming down. 

[nit2wynit]

2 minutes ago, DSchenk said:

My last statement for tonight, but strongly disagree with this. It makes all the difference. Measuring in £ doesn't make much sense at all

:O

Something is being Lost in Translation here.  I'm going to hold my hand up and say it's me haha.  We can pick it up another time if you like.

[dmedin]

19 minutes ago, DSchenk said:

The goal is to make £100k in profits starting with a £2k account

 

Anyone who has this aim would be better advised to spend £1 a week on playing the lottery, because they have a better chance of becoming a millionaire based off a random draw than making that much profit from spread betting.

[nit2wynit]

If i had 20k right now that I knew I could use to SB and could return 1% per day, 1k per week or 50k per year.  with 20k Margin, I could even use the FTSE and take 5 pts at a time, thus ensuring i'm not in too long and can take profit quicker.

Or is my math off?

Edited October 22, 2019 by nit2wynit

[DSchenk]

15 minutes ago, dmedin said:

Anyone who has this aim would be better advised to spend £1 a week on playing the lottery, because they have a better chance of becoming a millionaire based off a random draw than making that much profit from spread betting.

LOL

The chance of winning the National Lottery jackpot is 1 in 45,057,474.

The chance of winning in spreadbetting is (according to spreadbet providers) 1 in 20 (95% lose).
Ok, not everyone makes huge wins. I assume the chance of becoming a spreadbet millionaire is somewhere between 1 in 10,000 to 1 in 100,000. Certainly not in the millions

[DSchenk]

14 minutes ago, nit2wynit said:

If i had 20k right now that I knew I could use to SB and could return 1% per day, 1k per week or 50k per year.  with 20k Margin, I could even use the FTSE and take 5 pts at a time, thus ensuring i'm not in too long and can take profit quicker.

Or is my math off?

With £20k, you could do position sizes on the FTSE of 50. Means £50 per point. If you scoop 5 points of the FTSE every day, you make £250 per day, that's about £5k per month or £60k per year. You're math is correct.

I think this is an understatement of what spread betting can do.
Don't forget your purchasing power rises during the year as you make profits. From month 2 you can up your position size to 60, meaning you make now £300 per day, that's £6k that month. Month 3 you up size to 75. You make now £375 per day, £7.5k per month. And so on. Your account size reaches £1m in month 19. That largely matches with for example Ross Cameron's performance ($500 to $1m in 24 months, but taking profits out of the account above $100k)

Additionally, I think it should be more than doable to scoop more than merely 5 points per day. 10-20 points seems to be a realistic target.

[nit2wynit]

Agree with all above.  but 5 pts at a time x 4.

 

[Caseynotes]

99.999999% of traders lose money with this provider

[dmedin]

2 hours ago, DSchenk said:

The chance of winning in spreadbetting is (according to spreadbet providers) 1 in 20 (95% lose)

It's more like 99% or even 99.99%, and the number who can turn £2000 into £100,000 from spread betting in a single lifetime is probably as close to 0 as you can get.

It's a total mug's game.

Edited October 22, 2019 by dmedin

[dmedin]

40 minutes ago, Caseynotes said:

99.999999% of traders lose money with this provider

 

At least you're only losing £1 a week.  Practically all day traders lose considerable sums of money.  And when I say practically, it is about as close to all of them as you can get statistically (there will always be a few flukes).

Buy and hold a diverse portfolio of ETFs/index funds is your only chance of making ANY returns over the long run, pal.  That is the ONLY thing that's been proven in over 100 years of stock market research. 

[dmedin]

Sorry chaps but all I've got is my own experience and academic and statistical research versus your word.

I've never once seen any 'day trader' prove that they made decent returns over a sustained period.

Why is that?  You'd think they would be eager to show people how good they are so that people would pay to get their 'tips' and 'training courses'. 

[dmedin]

3 hours ago, nit2wynit said:

I've just made another £7 on the FTSE Drop.  My Minimum Stop was still 8pt but i was right at the top of the resistance point.  I got out right away to take a clear profit.

I suppose what I'm really commenting on is how we all see different things on screen and how we play it out.

 

 

Now let me ask you this: did you factor in the amount of time and effort required to earn £11.18?

Because time and effort are the things that never get considered in the risk/reward **** that these snakes in suits drone on about.

So you spend all afternoon charting and gambling with your money, and you come out of it with a twenty quid profit.  But you could have got a job somewhere and earned £20 per hour.  Less risk and more reward.

How fking stupid is that?

Edited October 22, 2019 by dmedin