NinjaTrader Strategies - Mechanical Trading General Discussions - Ask general questions, post charts and videos - backtesting, optmisation and strategy anlyasis
04/07/2010 16:51:40
Tom Leeson
Administrator
Posts: 661
|
I am not sure what you mean by “to try & go short when the market is trending long, within a trend channel bouncing off/violating the 10 ma,”
Do you mean that this is he better system – yes a new trader would think that… Lets just say I was saying look – here is another potentially viable system – not here is a replacement for the Range Trader. Time should be spent mastering the art of using the strategy analyzerr before you can really have an opinion of how the land lie on that one… Interpreting of results and whether there are tradable is the other skill.
You will learn over time that looks are irrelevant compared to:
Statistical evidence produced by optimisation and back testing – followed by an out of sample test is what is important.
Then on top of that risk management, money management, risk of ruin , + filters such as risk reward etc.
The key is to half the risk.
When you say that looks good you are a victim of recent bias and jump on it excited that you have found the holy grail – the holy grail is the trading framework not one system. The driver and the car + the team is what is important in F1 not just the engine in the car….
Traders will use their eyes to pull to what looks good and work on it – to try to avoid the loosing trade – sure that is what you have just said:
Looks better than the RANGE TRADER.
There are only 4 possibilities in mechanical trading and gambling systems:
Low winning frequency – string of small losses followed by an occasional large win
High winning frequency – string of small winners followed by an occasional large looser
More Opportunities with greater risk
Less Opportunities with smaller risk
This is not gambling though we can take advantage of the non random nature of the markets whereas in roulette you will loose –but the science behind the system is the same.
We will combine systems which are in themselves viable.
We need to add system together that do not duplicate functionality – but add on to it.
A Range breakout system + a Momentum price change system – is a good example.
The holy grail is many things all put together…..
RISK OF RUIN – are you adequately capitalised to avoid begin wiped out by a drawdown?
This is not calculated by 2.5% as you might think but by system expectancy, capital, allocation, cut off
The more £ the less likely it is to happen….
The risk of ruin is a study of how likely you are to reach the point of ruin – where you can no longer afford to place a trade.
EG. 10K on a 60% winning system is around 0.33.
20K is around 0.11
1 is total failure
0 is never failure
To avoid this we must diversify -
Part of it is diversification of portfolio, systems and settings
It achieves the following:
Less risk exposure, blah.
10 opposing systems and 5 assets will allow us to reduce risk by about 50% - but we cannot afford it. £250,0000 used to be a minimal sum.. but margins are cheaper and futures and forex let us revers positions unlike the monolitihic ouch i lost 75% of my equity mutual funds...and pension
We may be able to pull this off in FOREX pairs… with eg 8 pairs and very cheap transaction costs - we may actually be able to play it like a hedge fund on $1000 lol...
ANSWER TO YOUR COMMENT
Looks better than RANGE TRADER
Understand that the RANGE TRADER will perform baldy on days like that as is… it is a MEAN REVERSION SYSTEM – with a trend element – the entry ma…
If RANGE TRADER is making 0 when the FIB TRADER is making 1000s then they are the perfect pair if
When RANGE TRADER is Making 1000s the Fib Trader is making 0
Diversification needs to consist of opposites/ uncorrelated . negatively related systems and asset classes/assets….
Trend Following
Mean Reversion
Currency
Futures
Stocks
Bonds
Art
Antiques
Property
Loads of capital….:-)
or just some really great talent and learn one market and one system and discretionary trade it while using signals from robots as a heads up etc...
answered inspired by Perry Kaufman
edited by tom on 04/07/2010
--
MicroTrends tom
|
|
0
• permalink
|
04/07/2010 16:48:48
Tom Leeson
Administrator
Posts: 661
|
TRADER: "That FibTrader system looks better than the Range Trader and Trend Trader, it removes the need for the strategy to try & go short when the market is trending long, within a trend channel bouncing off/violating the 10 ma, have youany backtest results?"
--
MicroTrends tom
|
|
0
• permalink
|
Powered by AspNetForum 7.0.1.0
© 2006-2011 Jitbit Software