My algo trading daily performance

[vagar11]

can anyone make me understand how can a profit giving strategy in back-testing turn into loss giving in real time algo trading ??

provided that the strategy still worked in real time when traded ?? if we are not optimizing at all

Over optimized or a large drawdown in a effective strategy.

 

[princy01]

Over optimized or a large drawdown in a effective strategy.

suppose there are no indicators in a strategy , and its solely on price action and not any optimization is done , then ?

 

[ndtrader]

can anyone make me understand how can a profit giving strategy in back-testing turn into loss giving in real time algo trading ??

provided that the strategy still worked in real time when traded ?? if we are not optimizing at all

Happens all the time, but due to different reasons every time For starters:

1) How close to real life trading is your backtesting? Did you factor in slippage, bid-ask spread, liquidity, taxes etc.?

2) How exactly are you doing backtesting in the first place? Backtesting design is a big factor - if you are doing something like building your price action model on 90% data and testing it on 10% only, that's highly unrealistic.

Difficult to give a more meaningful answer without the specifics.

By the way, which broker & platform are you using for algo trading?

 

[AlgoAdv]

can anyone make me understand how can a profit giving strategy in back-testing turn into loss giving in real time algo trading ??

provided that the strategy still worked in real time when traded ?? if we are not optimizing at all

Profitable back tested strategy can make loss due to any of the following reasons:

1) Proper back testing was not done.
2) Slippage, commission, turn over charges not incorporated.
3) Fault in Strategy coding.
4) Change in market conditions.

 

[vagar11]

suppose there are no indicators in a strategy , and its solely on price action and not any optimization is done , then ?

If that's the case, you have tested the strategy on limited data. It should be tested for all types of days. Down, up, sideways, gap up , gap down, sudden reversal etc.

Personally, I test strategy for 4-5 years using programs. Even after that, I am not able to handle drawdowns and I end up going back to gambling. I hope I will get out of this habit and stop taking contrarian trades(Glenmark recently).

 

[vagar11]

If that's the case, you have tested the strategy on limited data. It should be tested for all types of days. Down, up, sideways, gap up , gap down, sudden reversal etc.

Personally, I test strategy for 4-5 years using programs. Even after that, I am not able to handle drawdowns and I end up going back to gambling. I hope I will get out of this habit and stop taking contrarian trades(Glenmark recently).

Atleast, I don't do averaging.

 

[mohan.sic]

can anyone make me understand how can a profit giving strategy in back-testing turn into loss giving in real time algo trading ??

provided that the strategy still worked in real time when traded ?? if we are not optimizing at all

where did you back test - on what application
what is Data time frame and period of testing

 

[princy01]

It was only a question

As of now no testing has been done but, I need 5 years of data before I start, need atleast nifty and bank nifty data for last 5 years minimum

 

[vagar11]

It was only a question

As of now no testing has been done but, I need 5 years of data before I start, need atleast nifty and bank nifty data for last 5 years minimum

https://drive.google.com/drive/folders/0B8e3dtbFwQWUZ1I5dklCMmE5M2M

 

[Blackhole]

@mechtrader Any update on your algo trades?