LIVE real money S&P500 daily mechanical trading

ordinaryguy76

Established member
Messages
618
Likes
16
Right, I've decided to "go for it" and start trading my mechanical system for S&P500 with real money.

To cut the long story short, the formulas are looking at open (previous day close), close, high and low over the previous three, four or five days and try to tell me whether I should buy or sell.

This was tested over the last 10 years of S&P data and the best formula (highest return compared to drawdown) was selected. Of course, historical profits can never guarantee future success. The biggest issue is if the stop loss (set in stone to 150 pips) gets triggered outside of S&P trading hours (currently 2.30pm - 9pm UK time), because my test data did not include out-of-hours movement.

Example:
Trade - Buy S&P @ 1260.0, SL 1245.0 (150 pips)
Order - Sell S&P @ 1244.0 (10 pips lower), SL 1259.0 (again 150 pips)

You can see results of 5 weeks' testing here:
http://www.trade2win.com/boards/tra...-semi-automated-s-p-daily-trading-system.html

What I like about this system is this:

1. it completely ignores fundamental analysis and 99% of technical analysis (it does not use any known indicators, but you can argue that my formulas are indicators on their own). You could say it's totally random :)

2. all trades are opened / closed at 9 pm, so you only need 0.5h per working day. You don't have to spend hours and hours in front of the screen and jump up at some news from EU or else. I am working full-time and have two wonderful little kids, so 0.5h daily is a lot of time for me.

3. you know that you can't lose more than 150 x stake per trade and 300 x stake per day. It's very easy to limit your risk to 1.5%, 3%, 6%, whichever you prefer. I will start initially with 3% risk per trade, but this will increase in 2 months' time, if I am successful.

4. 50% success rate is more than enough. Why ? Because you can only lose X amount of money on a losing trade whereas a winning trade can give you (a lot) more.
It's that 50% you're aiming for.

I am going to post live trades. If you have guts, you can follow them. I will try to make it easier for you and tell you IN ADVANCE how I will trade at 9 pm.

Oh my, what if it does not work ? Well.... I will come back to this forum with a different nickname perhaps ? hehehehe
 
ok, so I've opened a fresh account with Finspreads, because they allow you to trade S&P500 with 10p per point for 8 weeks.

My initial bankroll is £1000 (don't worry, I can afford losing it all), but I have only deposited £500 at the moment. Why should I keep too much of my money with some institution that I know almost nothing about ?

Finspreads offer "US SP 500 Rolling Spread". For some reason it's around 15-20 points higher that the actual S&P500. I will try to ask them about it at some point. Just wondering how this might affect my trading.

Initial deposit: £500
First trade:
Sell £0.20 S&P @ 1258.5, SL 1273.5
No opposite trade for now, as this trade is a bit unusual (there is bank holiday on Monday).
 
I wish you good luck with this.
My initial observations are that the stop is too wide for this type of trade to work as the importance of the data is nullified because of the wide stop.what are the expected returns on a £10,000 account in 12 months? Thanks
 
the s&p doesn't even move in "pips". 1258 to 1259 isnt 10 pips it is 1 handle or 4 ticks. might wana learn your market..
 
the s&p doesn't even move in "pips". 1258 to 1259 isnt 10 pips it is 1 handle or 4 ticks. might wana learn your market..

You are describing the emini contract, ES.

The full S&P500 futures contract (SPH) trades at a minimum of .10 index points = $25, so 15 points x 10 ticks per point = 150 ticks.

I think he means the full contract but he's trading at 10p stakes.

Peter
 
geez i never new that! lol. was kinda presuming with his 500 quid account he wasnt trading big spoos in the pit!

also you wil be lucky to get filled on .10's you will almost certainly be fille don 20s 40s 60s 80s etc..locals use the 10s to flip
 
"It is not cheap to develop automated systems , mine already cost over $1m. " oildaytrader

LOL. you sig line still makes me laugh. Now that you're back can we get oily back too? I miss the two of you arguing over his stupidity!

Peter
 
Hello, I'm not even trying to say that I have a lot of experience in trading. At the moment, I'm using a spread betting provider. They don't use lots, handles etc. In terms of "filling", it does not work this way, either. You just buy / sell at the price offered by spread betting company.

Thanks for the encouragement. Wishing everyone all the best in the New Year !
 
I was looking in disbelief at my trade and realised something quite interesting and helpful for newbies. It seems that finspreads allow you to trade S&P per 1.0 (!) whereas all other spread betting companies that I know of, trade S&P per 0.1

This means that trading 10p stake with Finspreads (per 1.0) is like trading 1p with everyone else. Well, it also means that my stake for initial trade was ten times too small, same with the risk.

Anyway, knowing this, Finspreads would be my first choice for spread betting newbies hands down.
I will say it again: you can open an account with £100 and trade 1p per 0.1 of S&P500 ! Everywhere else, it's a minimum of 50p per 0.1, e.g.
- worldspreads - best so far for spreads, hedging trades and opening offer (the only cashback that I had to use :) )
- etxcapital - best so far for quickest financial operations like deposit / withdrawal. Has issues with trailing stops, though
- spreadco - the only spread betting company that shows volume on their weird charts
- capitalspreads - even more risky - min. £1 per 0.1. Also, look out for slippage on your stops.

My next trade on Finspreads will be then for £2 (i.e. 20p per 0.1 of S&P)
 
As for the 12 months' return.

I don't really know. What I saw when backtesting was the consistency of profits over longer periods of time. I can mean return of 30% one year and then -5% another year. I am still refining my formulas to reduce losses and increase winnings. One of the methods is to keep risk on the same level, i.e. once the account value goes up by 10%, increase stake by 10% and when it goes down, decrease stake accordingly. We'll see.
 
wow, this is what I call start with a bang !
Actually, I was kind of expecting markets to go up on the first trading day in the new year. This is what happened over the past few years. Still, this system pointed to a sell, so I did a sell...

Now, my trade was stopped out @ 1282.4 for a total loss of £4.78 only, thanks to the fact discussed in post #9 (stake / risk 10 times smaller than intended). This was lucky.

No more trades before 9 pm.
9 pm trade will be a buy, regardless of where S&P ends tonight.

I might increase my stake to £3, instead of the planned £2. I just noticed that finspreads have a bonus offer where they add £100 to your account if you make 3 trades £3 each.
 
Last edited:
Phew, I'm still trying to get used to Finspreads interface.

New trade entered:
03/01/2012 21:00
Buy £3 (that's 30p per 0.1) S&P500 @ 1277.9, SL 1262.9
Sell order £3 S&P500 @ 1261.9, SL 1276.9
 
Hmm, weird stuff. My trade was kind of closed and reopened (rolled ?) around 3am. The only issue is, I lost £0.60 along the way. Currently, my trade shows opening price of 1277.7. I don't like this and I've emailed Finspreads for explanation. Such a thing never happened when using other providers - rollover never changed opening price of my trades.
 
at the moment, it seems you should be doing completely the opposite to what I'm doing :)
Hopefully, my loss will be reduced later on.
 
By the way, I'm 95% sure 9 pm trade will be UP, unless S&P really starts flying now (unlikely)
 
hmm, never say never, huh ? My trade might still turn positive. If it does, it will be a sell tonight.
 
This was tested over the last 10 years of S&P data and the best formula (highest return compared to drawdown) was selected. Of course, historical profits can never guarantee future success. The biggest issue is if the stop loss (set in stone to 150 pips) gets triggered outside of S&P trading hours (currently 2.30pm - 9pm UK time), because my test data did not include out-of-hours movement.

A couple of observations from what you've said in your opening post. Firstly if you're trading out of US market hours, but your sample data didn't include it, then the results you got are most likely not very accurate. As the S&P 500 futures (ES) move around a fair bit out of hours - especially during the European session starting from about 7am UK time, so to get a more accurate sample you probably want to back test using the ES futures data instead.

Also the ES roughly currently has a 200 day average true range of 26.36 full points (263.6 pips using your example), whereas the actual S&P 500, (so the US session only) has a 200 day average true range of 22.10 (210 pips using your example). So what made you decide on the set in stone 15 point (150 pip) stop loss as the average daily movement of it is much higher?

Finally remember to factor in that the spread is doubled out of hours on Finspreads to 0.8 points.
 
OK, trade closed by me for -£1.20. Adding -£0.60 realised overnight gives -£1.80.
Opposite trade cancelled.
Two more trades to apply for the bonus.

Account balance is £493.14.

I don't really like such flat days and am concerned whether I made the right decision with the next trade.

Here it is anyway:

04/01/2012 21:00
Buy £3 (that's 30p per 0.1) S&P500 @ 1277.7, SL 1262.7
Sell order £3 S&P500 @ 1261.7, SL 1276.7
 
A couple of observations from what you've said in your opening post. Firstly if you're trading out of US market hours, but your sample data didn't include it, then the results you got are most likely not very accurate. As the S&P 500 futures (ES) move around a fair bit out of hours - especially during the European session starting from about 7am UK time, so to get a more accurate sample you probably want to back test using the ES futures data instead.

You are absolutely right. I would much rather test on 2.30 pm open to 9 pm close as well, but I don't have access to such data going back years. I was able to get around 1 month only of detailed S&P futures data. If you can point me in the right direction, I will be grateful.

Also the ES roughly currently has a 200 day average true range of 26.36 full points (263.6 pips using your example), whereas the actual S&P 500, (so the US session only) has a 200 day average true range of 22.10 (210 pips using your example). So what made you decide on the set in stone 15 point (150 pip) stop loss as the average daily movement of it is much higher?
My stop loss was one of the variables. I kept changing it until I got satisfactory results considering drawdown and daily S&P range. I think that the highest profits were with even smaller SL around 120 pips, but I decided on 150 for various reasons, like variability of daily results, drawdown etc.
I like bigger daily ranges (one way) because my losses are limited to 150 (and if the market goes further, then the trade is reversed for profit) whereas profits can run much higher.

Finally remember to factor in that the spread is doubled out of hours on Finspreads to 0.8 points.
I didn't know about this, thanks. I used other companies and they did not change their spreads on S&P after hours. FTSE, WS, currencies - yes, but not S&P500. Weird.
 
Download yourself a free copy of Ninjatrader from: NinjaTrader stock, futures and forex charting software and online trading platform. and then I used the free futures zen-fire demo from Futures Trading Broker | NinjaTrader with Zen-Fire | Online Futures Trading | Mirus Futures which lasted about a month and gave me about 5 years of intraday tick data for the ES to experiment with. Ninjatrader's backtester takes some getting used to, but you should be able to get it working after you've watched a few of their webinar videos on the Ninjatrader site or go over to Big Mikes trading forum as is 90% Ninjatrader users.

Hope that helps
 
Top