I've been there. Thinking I'll wait untill breakeven to exit only to lose more. I think if you decide right it's been Xweeks, the bet hasn't done what I expected, I should just exit and move on. It'll be sods law that as soon as you exit the price goes in your favour but you just need to exit. It's hard to sdo but once you've made the descision you want to exit you just need to get out.

The only other option is to leave the bet alone and let it pan out whatever way that might be. Move stops according to the strategy and stick to it. So basically only exit the trade when the stop is hit. The only problem with that is it can potentially do nothing forever, and our trading funds will be tied up which we could use elsewhere.

Or if how you said we have a stock stuck between 0% & -2% of our open price and we decide we want to get out of the trade we could set a limit order to get out at 0% and a stop to get out at -2%. That way we know the best we will do is breakeven the worst will be -2%, which is better than -5%.

A few ideas not really sure which is the best!

_________________
May the markets be with you!

Harry,
The Spread betting beginner

If we are using trailing stops then there is no need to set a time limit on a trade.
OTL, is the reason you decide to sell a trade that is going nowhere after 10 days purely because of overnight finance charges?

_________________
Spread Betting Trader
http://spreadbettingtrader.blogspot.com/

Ok so I think this topic has gone off the boil a little (my fault, times tight as always). So in the interest of moving it forwards I propose we use a 2.5xATR(10) trailing stop and this will govern when to exit the trade and the length of a trade. We will not have a time limit on the trade and it could potentially stay open for ever.

I think the problem is none of us knows what to do for the best. If we did then we'd probably all be millionaires. I don't think we have stumbled across the 'Holy Grail' of trading strategies here but then again I personally don't think any such 'Holy grail' strategy exists. A trading strategy is a very personal thing and a strategy that makes one person millions will not necessarily do the same for some one else. Let's take BigAl as an example (sorry BigAl, I don't mean to single you out). Big Al is trading a strategy by Vince Stanzione. If you Google Vince Stanzione you will find no end of people that say he's a scammer and that his strategies don't work. But look at BigAl turning £800 into £2400 in less than a year trading a strategy from him. I think it's as much about the person as it is the strategy and obviously BigAl has the spread betting knack which suits the strategy he trades.

The strategy that we are in the process of developing will work for some people, who knows it could even make someone a millionaire, but there will be others that simply cannot make it work and they should continue their search for their own 'Holy Grail' trading system.

So to summarise:
• Setup - DONE
• Entry triggers - DONE
• Stop placement - DONE
• % of funds to risk - DONE
• Position size - DONE
• When to move stops - DONE
• When to exit the trade - DONE
• Length of the trade - DONE
• Trade parameters to log - in Progress


All we need now are the trade parameters to log. To be honest I'm very much looking at off the lip to take the lead on this part. Also off the lip you did some back testing a while ago and now could be a good time to run the back tests again to see if it gives us an edge.

_________________
May the markets be with you!

Harry,
The Spread betting beginner

On the trading parameters, I think we need to record, everything about the entry that we have decided as being significant, e.g. what the value of RSI actually waas. Similarly with the stop placement, was the intial stop, was it decided on the lowest bar or on the limiting aTR value. Also we need to record what the intial stop was, I have been doing this in order to work out the Van Tharp R for each trade. Malcolm Prior recommends this and it seems to be a good way of assessing a strategy.

On the back testing, I will give it a go, however I am a bit busy at the moment so it might be a week or two before I can get it coded because it has got a bit more complicated now.

Off the Lip has very kindly run the back tests for the new spread betting strategy that we have created. I wasn't sure where I was going to add these so what I have done is created several Google docs spread sheets to display the information that Off the Lip has sent me. The links to the spread sheets are below and off the lip has also provided comments to get the discussion going.

I want to take this opportunity to say a big thank you to off the lip for taking the time to run the tests and add some additional comments. It's appreciated as always.

The Long Tests

Long Tests - Original Values
Long Tests - Limits Reduced
Long Tests - Infinite Limits

1: The strategy only generated 44 samples over my test period which is enough to be statistically significant but is still not a large sample and thus care should be taken in drawing any conclusions.

2: The original settings show a slight profit but not enough to cover the spread of most of the stocks.
Ave all 0.516328484
Ave win 4.691895721
Ave loss -3.29614595
No win 21
No losses 23
               
3:  If you set up a filter on profit and look at the winners and then look at the average of Minp/Initial stoploss you see that the average is very nearly 1.03. What this means is that on the winners the minimum price reached was on average 3% above where we put the initial stop loss. This implies that we put could raise the initial stop loss and expect to get an improvement. Thus I ran two more tests one using just the last 3 bars ( the infinite tab) and secondly using a setting of X and y of 0.5 and 1.5.
Infinite x and y (ie using last three bars only)
Ave all 0.651987588
Ave win 4.233707373
Ave loss -1.603169313
No win 17
No losses 27
 
X=0.5 y=1.5
Ave all 0.689986606
Ave win 4.456220727
Ave loss -1.917406248
No win 18
No losses 26
 
What this does show is that the results are easily changed by changing the stoploss setting.
With all of these test I didn’t change the trail at 2.5ATR. I suspect that since we seem to get improvement by setting our stop closer than 2.5 ATR that we would also get improvements by trailing closer than 2.5 ATR at least up to breakeven.

4. If you take the filter off, and look at the Maxp/ Exitprice this shows an average of nearly 1.05. This means that we end up giving an average of 5% back from the maximum price reached before we exit. I suspect we could improve this. Note also that all but one of them are greater than 1, which means that are possibilities of improving the exit method even for the losers.

_________________
May the markets be with you!

Harry,
The Spread betting beginner

The Short Tests

Short Tests - Dif Limits and Trail
Short Tests - Original
Short Tests - Infinite
Short Tests - Dif Limits
Short Tests - Trail at 1.5

1: The back test only triggered 29 cases over the test period so this is only a small sample and is not really large enough to be significant.

2: Over the period of the test from June last year to present the market has been in Bull mode so it is not surprising that a "short " strategy is struggling. The good news is that is doesn't seem to lose a lot. I do have other periods that I can run the test over, but there is a bit of a limitation of my system: I tend to keep data file is chunks of about 11 months of data, this is so that I can test over 1 year of data with 1 month run in for moving averages. Since our strategy requires a 200 day moving average which is about 10 months of data in theory I can only test accurately on 2 months of data. I actually  started the test after 6 months of historic data reckon that the 200 day EMA will b very close to the 200 day ema started only 100 days before.

 3: The percentage of winners of only 26% is low, which means that there is a strong likelihood of there being long lines of losers, this test on the "original settings" starts off with a run of 9 losers with a total loss of 37% . One needs a strong psychology to trade a strategy like this. 

4: The winners when they arrive are big winners!! This also means that this strategy is likely to have very inconsistent results since it relies on how big the winners are and they don't happen very often.

5: Because the strategy does produce big winners I tried tweaking the strategy to reduce the loss on the losers by playing with the initial stop loss settings and the trail factor. The tabs show the following differences:
Original : Obviously the original settings
Inifinite: Intial stop set at High of last 3 bars
Dif limits: Initial stops set at Hi of last 3 bars, limits to between 0.5 and 1.5 AVTR
Trails at 1.5: Initial stops set at Hi of last 3 bars, limits to between 0.5 and 1.5 AVTR, but the stop trailed at 1.5 AVTR
Dif limits and trail: Initial stops set at Hi of last 3 bars, limits to between 0.3 and 0.8 AVTR, but the stop trailed at 1.2 AVTR

_________________
May the markets be with you!

Harry,
The Spread betting beginner