There are 10 winners & 51 losers - psychologically quite a tough system to trade with I think. Also, I doubt whether in practice 3 / 5 pip losers are realistic, taking into account the bid/ask spread, even if you use stop orders. I'd be interested to see the result using a longer bar time with wider stops.
(a) allowing the period within which the band is established for breakout, and the time of closing of trade are read in from the sheet itself
(b) to record time of entry of trade
(c) to clear old data
(d) execute on CTRL shift Q
To modify time of exit from 21:00 it is important to block trades that start after the closing time.
Pretty much confirms other results
An interesting diversion is to only enter a trade when it has moved a significant distance already:
See the attached run data - only hopping on to a trend if it has moved 100 ticks in either direction from the position at 8.0am - interestingly this gives the highest points/trade - rarely suffers a reversal ( and no whipsaws) although does not trade every day
Johnny : any chance you can post the data back to september to allow more rigorous testing?
Excellent macro! I was able to squeeze out a couple more points per trade by reducing profittarget to 0.012 to filter out eod reversals. Average trade 52.92 over 13 trades. An extended back-test would be interesting.
I've been collecting more backdata, mainly since 20th Feb when JT's ran out (actually been tracking one of his other freebies since then, both on spreadsheet with backdata and CS simulation account and getting total correlation), and since 20th Feb it's only netted another 180 points.
It seems that this period, start of Jan-20th Feb works really really well for all of Jonny's strategies, but certainly this one and the other I've been looking at (think it's called "is this one spot on?") seem to go off the boil a bit since then. In fact, if you split the results of this strategy into months it goes Jan 452, Feb 267, March (1st half)115.
Don't know what this means, but the other strat. seems to fail completely when I look at Dec03's data (which interestingly, if you look at a chart of daily candlesticks, is almost a straight line line upward trend all month!). I've put 1min daily backdata together twice now (what a flippin' tedious job that was!) and still am not getting the returns from early this year. There just don't seem to be enough hours in the day for me!
Has anyone else done any more extensive backtesting?
Has anyone got a good set of 1min or maybe 5min data that we could all share for backtesting and drawing out these relatively simply and potentially lucrative strategies?
Could we co-ordinate some analysis here, instead of all doing it separately? Surely that's one of the advantages of this kind of forum?
I fear we may be lulled into a false sense of security with that 6 weeks of data ...
I'm about to start marketing my new system. 60 min chart for crude oil. Buy when 14 RSI hits 20. Sell when it hits 80.
We all know that it's not sustainable, but on the basis of the attached chart I should be able to persuade some suckers to part with their money :cheesy:
But seriously, this instrument often seems to start a trend in the first hour each day (particularly if there's some news from Europe/Middle East in the morning), so it could be a candidate for JonnyT's breakout system.
I did some back-of-a-fag-packet calculations that showed about 400 points profit over the last five weeks using a breakout from 15:00-16:00, exit at 19:30. Anyone know where I can get some (preferably free) historical data to do a proper backtest :?:
which type of oil is the chart.brent crude or sweet light.you can trade this as a spreadbet.both can be traded at capital spreads.they also have charts/15 min delay,data supplied by netdania.i will find out the spread and post it later