Would you trade this system?

A very LAZY End of Day system, between 5-7 trades triggered a year.
Average holding time for a winner over 400 days.

Id be very grateful to get any feedback and construstive criticisms.
If you wouldnt trade it -- why not?
And in which ways can it be improved?

There is a 2-year period in which the system made no money at all (1995-1997 see monthly return chart) -- Is this acceptable?

The testing period is 10 years, 1992 until 2002, over the entire ASX market.

Monte Carlo Report

Trade Database Filename
C:\TradeSimData\BlueprintMachinewithOoooomphLATEST.trb

Simulation Summary
Simulation Date: 13/09/2007
Simulation Time: 2:29:09 PM
Simulation Duration: 193.33 seconds

Trade Parameters
Initial Capital: $30,000.00
Portfolio Limit: 100.00%
Maximum number of open positions: 100
Position Size Model: Fixed Percent Risk
Percentage of capital risked per trade: 2.00%
Position size limit: 100.00%
Portfolio Heat: 100.00%
Pyramid profits: Yes
Transaction cost (Trade Entry): $44.00
Transaction cost (Trade Exit): $44.00
Margin Requirement: 100.00%
Magnify Position Size(& Risk) according to Margin Req: No
Margin Requirement Daily Interest Rate (Long Trades): 0.0000%
Margin Requirement Yearly Interest Rate (Long Trades): 0.0000%
Margin Requirement Daily Interest Rate (Short Trades): 0.0000%
Margin Requirement Yearly Interest Rate (Short Trades): 0.0000%

Trade Preferences
Trading Instrument: Stocks
Break Even Trades: Process separately
Trade Position Type: Process all trades
Entry Order Type: Default Order
Exit Order Type: Default Order
Minimum Trade Size: $500.00
Accept Partial Trades: No
Volume Filter: Reject Trades if Position Size is greater than
10.00% of the maximum traded volume
Pyramid Trades: Yes
Favour Trade Pyramid: No
Start Pyramid at any level up to level: N/A
Maximum Pyramid Level Limited to: N/A
Maximum Pyramid Count Limited to: N/A

Simulation Stats
Number of trade simulations: 50000
Trades processed per simulation: 668
Maximum Number of Trades Executed: 73
Average Number of Trades Executed: 65
Minimum Number of Trades Executed: 57
Standard Deviation: 8.00

Profit Stats
Maximum Profit: $532,904.37 (1776.35%)
Average Profit: $507,873.68 (1692.91%)
Minimum Profit: $481,811.58 (1606.04%)
Standard Deviation: $23,488.87 (78.30%)
Probability of Profit: 100.00%
Probability of Loss: 0.00%

Percent Winning Trade Stats
Maximum percentage of winning trades: 45.61%
Average percentage of winning trades: 40.61%
Minimum percentage of winning trades: 35.62%
Standard Deviation: 5.00%

Percent Losing Trade Stats
Maximum percentage of losing trades: 64.38%
Average percentage of losing Trades: 59.39%
Minimum percentage of losing trades: 54.39%
Standard Deviation: 5.00%

Average Relative Dollar Drawdown Stats
Maximum of the Average Relative Dollar Drawdown: $6,256.97
Average of the Average Relative Dollar Drawdown: $5,457.29
Minimum of the Average Relative Dollar Drawdown: $4,614.18
Standard Deviation: $794.30

Average Relative Percent Drawdown Stats
Maximum of the Average Relative Percent Drawdown: 3.5568%
Average of the Average Relative Percent Drawdown: 3.0507%
Minimum of the Average Relative Percent Drawdown: 2.5184%
Standard Deviation: 0.5034%

Maximum Peak-to-Valley Dollar Drawdown Stats
Maximum Absolute Dollar Drawdown: $49,609.25
Average Absolute Dollar Drawdown: $39,053.76
Minimum Absolute Dollar Drawdown: $28,120.89
Standard Deviation: $10,491.17

Maximum Peak-to-Valley Percent Drawdown Stats
Maximum Absolute Percent Drawdown: 16.0116%
Average Absolute Percent Drawdown: 15.2180%
Minimum Absolute Percent Drawdown: 14.2264%
Standard Deviation: 0.8054%

Profit Summary
Profit Status: PROFITABLE
Starting Capital: $30,000.00
Finishing Capital: $559,826.44
Maximum Equity/(Date): $538,872.46 (10/09/2002)
Minimum Equity/(Date): -$2,827.01 (9/03/1993)
Gross Trade Profit: $630,743.44 (2102.48%)
Gross Trade Loss: -$100,916.99 (-336.39%)
Total Net Profit: $529,826.44 (1766.09%)
Average Profit per Trade: $9,295.20
Profit Factor: 6.2501
Profit Index: 84.00%
Total Transaction Cost: $5,016.00
Total Slippage: $0.00
Total Trade Interest: $0.00
Daily Compound Interest Rate: 0.0806%
Annualized Compound Interest Rate: 34.2018%

Trade Statistics
Trades Processed: 668
Trades Taken: 57
Partial Trades Taken: 0
Trades Rejected: 411
Winning Trades: 26 (45.61%)
Losing Trades: 31 (54.39%)
Breakeven Trades: 0 (0.00%)

Largest Winning Trade/(Date): $122,650.19 (3/07/2002)
Largest Losing Trade/(Date): -$9,675.20 (28/06/2002)
Average Winning Trade: $24,259.36
Average Losing Trade: -$3,255.39
Average Win/Average Loss: 7.4521

Trade Breakdown Long and Short Trades Long Trades Short Trades
Normal Exit: 25 (43.86%) 25 (43.86%) 0 (0.00%)
Protective Stop: 29 (50.88%) 29 (50.88%) 0 (0.00%)
Open Trade: 3 (5.26%) 3 (5.26%) 0 (0.00%)

Total Trades: 57 (100.00%) 57 (100.00%) 0 (0.00%)

Trade Duration Statistics Winning and Losing Trades Winning Trades Losing Trades
Maximum Trade Duration: 1167 (days) 1167 (days) 175 (days)
Minimum Trade Duration: 2 (days) 10 (days) 2 (days)
Average Trade Duration: 217.00 (days) 425.19 (days) 42.39 (days)

Consecutive Trade Statistics
Maximum consecutive winning trades: 4
Maximum consecutive losing trades: 8
Average consecutive winning trades: 2.00
Average consecutive losing trades: 2.21

Trade Expectation Statistics
Normalized Expectation per dollar risked: $4.36
Maximum Reward/Risk ratio: 44.01
Minimum Reward/Risk ratio: -1.93
Average Positive Reward/Risk ratio: $10.73
Average Negative Reward/Risk ratio: -$0.98

Relative Drawdown
Maximum Dollar Drawdown/(Date): $28,881.95 (18/04/2000)
Maximum Percentage Drawdown/(Date): 14.6100% (18/04/2000)

Absolute (Peak-to-Valley) Dollar Drawdown
Maximum Dollar Drawdown: $28,881.95 (14.6100%)
Capital Peak/(Date): $197,642.89 (3/03/2000)
Capital Valley/(Date): $168,760.95 (18/04/2000)

Absolute (Peak-to-Valley) Percent Drawdown
Maximum Percentage Drawdown: 14.6100% ($28,881.95)
Capital Peak/(Date): $197,642.89 (3/03/2000)
Capital Valley/(Date): $168,760.95 (18/04/2000)


Performance Summary Report

Trade Database Filename
C:\TradeSimData\BlueprintMachinewithOoooomphLATEST.trb

Trade Statistics Long and Short Trades Long Trades Short Trades
Trades Taken: 57 57 0
Total Net Profit: $529,826.44 $529,826.44 N/A
Average Trade Profit: $9,295.20 $9,295.20 N/A
Maximum Trade Profit: $122,650.19 $122,650.19 N/A
Minimum Trade Profit: -$9,675.20 -$9,675.20 N/A
Break Even Trades: 0 0 0
Winning Trades: 26 26 0
Losing Trades: 31 31 0
Profitable Trades: 45.61% 45.61% N/A
Losing Trades: 54.39% 54.39% N/A
Average Winning Trade Profit: $24,259.36 $24,259.36 N/A
Average Losing Trade Profit: -$3,255.39 -$3,255.39 N/A

Charts attached.

Winning trades 26. Losing trades 31.
how is this a good system?

Hi Nizar

Thanks for posting. Personally I wouldn't trade this system for a number of reasons.

1. Too few trades to be considered statistically robust.
2. The largest winning trade is about 20% of profits from one trade. I prefer to keep the largest winner to a smaller percentage of the total profits.
3. The closed trade (MaxDD) drawdown is higher then I'd prefer personally for a low frequency trading system. I prefer a MaxDD under 10% for my weekly based systems which trade about 25-50 times per year. I like low drawdown system so I can use leverage to bump up the returns.
4. Long periods of no profit or poor returns may make the system hard to stick with. Capital is tied up and being unproductive for long periods.

Things I like are

1. Frequency does not equal profitability as your results show.
2. Low frequency means low commissions
3. Is it a long or short side system?

It's a start and ceratinly something to keep working on. Good luck.

Cheers

Shane

Shane Baker said:

Hi Nizar

Thanks for posting. Personally I wouldn't trade this system for a number of reasons.

1. Too few trades to be considered statistically robust.
2. The largest winning trade is about 20% of profits from one trade. I prefer to keep the largest winner to a smaller percentage of the total profits.
3. The closed trade (MaxDD) drawdown is higher then I'd prefer personally for a low frequency trading system. I prefer a MaxDD under 10% for my weekly based systems which trade about 25-50 times per year. I like low drawdown system so I can use leverage to bump up the returns.
4. Long periods of no profit or poor returns may make the system hard to stick with. Capital is tied up and being unproductive for long periods.

Things I like are

1. Frequency does not equal profitability as your results show.
2. Low frequency means low commissions
3. Is it a long or short side system?

It's a start and ceratinly something to keep working on. Good luck.

Cheers

Shane

Click to expand...

Thanks for your feedback Shane.
Yes points 1,2 and 4 are of concern to me as well.

Its a long only system.

mb1 said:

Winning trades 26. Losing trades 31.
how is this a good system?

Click to expand...

You have a lot to learn, my friend.

Take a look at average win/average loss, and also see the expectancy of the system.

I didnt ask if it was a good or bad system, rather, i wanted to know if you would trade this system, and also i asked for some constructive criticism.

If you are looking for 100% winners before you trade a system then you will be looking for a long time.

Another thing iv noticed with this method is that im getting stopped out way too much, with 51% of trades hitting the protective stop.

Any ideas why this is the case?

The stop i thought was quite loose at 3*ATR(10).

The exit is exactly the same as tech/a.
180 day ema of the lows.

nizar said:

Another thing iv noticed with this method is that im getting stopped out way too much, with 51% of trades hitting the protective stop.

Any ideas why this is the case?

The stop i thought was quite loose at 3*ATR(10).

The exit is exactly the same as tech/a.
180 day ema of the lows.

Click to expand...

Depends on the entry.

3 ATR's is conceptually much tighter for a breakout system than it is for a pull back system.

nizar said:

The stop i thought was quite loose at 3*ATR(10).

Click to expand...

I actually thought this was quite tight. What happens if you use a 10% of position size fixed initial stop, just for comparison sake?

nizar
Besides the reasons that Shane has outlined above I wouldn't trade it because it is based on daily charts - I cannot risk not making a trade because I am tied up in a meeting, out sailing etc. Also a daily system cannot handle as much money as a weekly system - I can buy / sell over 5 days rather than 5 hours. But enough of my biases!

Why have maximum trades equal to 100? Drawdown is quite high since it is closed trade drawdown. What does the equity curve look like?

On a daily chart I would think that 3*ATR as a maximum stop loss is quite tight for a long term system. I might use that for a weekly time frame. Why not optimise the multiplier and see what the 3D chart looks like - use a dummy parameter to get the 3D chart and also do a mini monte carlo.

regards
stevo

Nizar
Doh - I found the equity curve!

There are some big winners - it looks like 5 or 6 trades make up most of the dollar profit.

stevo

theasxgorilla said:

I actually thought this was quite tight. What happens if you use a 10% of position size fixed initial stop, just for comparison sake?

Click to expand...

ASX.G.

What do you mean exactly?

Allocate 10% of equity to each position?
Or use 10% of the purchase price as a stop-loss (like T/T) instead of the 3*ATR stop?

Stevo thanks for your thoughts.
Yes it needs alot of work.

Hi Nizar

Why not run the system with no initial stop and have a look at the results? Try plotting the R multiple return against the initial stop loss (use the trailing stop loss as the initial stop) as a percentage in a scatter graph (see example) and look at whether there are clusters that might suggest a logical stop placement in % terms. I still am concerned that your entry criteria are too tight and there aren't enough trades for a proper analysis.



Cheers

Shane

Nizar,

Out of interest would you post the Trade Log in an Excel spreadsheet?

The reason for the request is to have a gander at the value of the Initial Stops in the various trades.

Also, have you coded intra-day or next day exits (for both Initial and Profit Stops)?

Cheers,
rnr

Greetings --

Are the results being discussed in-sample or out-of-sample?

Only out-of-sample results have any value in predicting the future behavior and profitability of the system.

Howard
www.quantitativetradingsystems.com

Shane -- good idea, im working on it.

rnr -- the trade log is coming.

Howard -- in-sample. I have left aside 5years of data untouched to be used as an out-of-sample data, but im waiting until iv completed your book before i touch it. By the way, i received your book on Tuesday. Thanks. Its next on my list, after Iv finished the book by your best friend Bob Pardo LOL.

Only out-of-sample results have any value in predicting the future behavior and profitability of the system.

Click to expand...

Howard,
Whilst I respect that out of sample testing makes sense in certain scenarios, I really don't think its the only way to do things as you portend. As an example, a system that has not been optmised, has not been curve fitted and operates with the same parameters on every symbol, why will out of sample be any better than in sample in this case?

Nick Radge said:

Howard,
Whilst I respect that out of sample testing makes sense in certain scenarios, I really don't think its the only way to do things as you portend. As an example, a system that has not been optmised, has not been curve fitted and operates with the same parameters on every symbol, why will out of sample be any better than in sample in this case?

Click to expand...

Hi Nick --

Sorry to disagree, but I think every system must be tested out-of-sample. I believe that it is the only way.

In-sample results are always good. We do not stop designing until they are good.

At any rate, if a person develops a system without out-of-sample testing, then out-of-sample testing begins tomorrow, with real money.

I prefer to do the testing without the exposure to losing real money if I have made a mistake in the development process.

Thanks,
Howard