Baffled by Backtesting

without knowing more about which instrument your trading, the backtesting/walk forward method etc and the specifics of the strat its impossible to offer any worthwhile help.

what i can say though is that its very easy to make a sexy looking equity curve in a backtest, only for it to fall over in live trading. For me, the main reasons when this has happened has been due to unrealistic slippage or unrealistic execution assumptions. the avg trade is $17 ($116028/6801 right?), so that doesn't leave a lot of room for any errors with these. if your using stop/limit orders then the fill assumptions may be a problem., ie the price hits the limit/stop price and the backtesting software assumes a fill, whereas in real life the you wont get filled.

the best thing you could do would be to live test for a few months, but that's probably not what u want to hear :)

ps welcome to the dark side lol

W.r.t. slippage / commissions, these are included and are what I consider aggressive (as in higher than I think I will experience - I have traded them before). The strategy uses market orders that can be placed in advance (i.e. it's not "signal triggered -> straight to market" it's "if X, Y and Z are true place stop market at 1000 and wait until one of X,Y or Z is false or the order is filled".

I know that forward testing can be the only real arbiter of whether the strategy has any merit, but as I said earlier it means changing the setup I already have slightly.

As for the dark side... well up until a thread about random entries here, I was absolutely convinved that a mechanical strategy couldn't work... that is still sort of my defauly position too; but I've got this backtest and I can't put it to bed just yet.
 
mrg. did you ever live test this? would be interest to hear any conclusion about the system you have come up with.
 
I ended up buying over 4 years of tick data (spoos) to try it on that, and the results didn't match the other backtests I had done (tick data is necessary for the strategy to execute properly).
 
tick data eh :) thats another level of head scratching. v difficult to reliably backtest with (in my experience)
 
what i meant was that with tick data there is such vast amount of data, it makes testing errors much more likely and difficult to identify. also with such massive amounts of data, running any kind of opt software is so slow. for my experience any strat that relied on a few ticks profits has fallen down in live trading due to the reasons post previously. of course it v.possible imo, just requires another level of process knowledge, software/programming skills and top hardware to crunch the numbers.
 
3 to 4 years back tests is not enough.Your results on first post sound okay for mechanical strategies i.e 2 to 3 pips/ticks per trade.

Back tests give instant fills , real market has slippage and many unfilled trades if orders are limit orders.

If market conditions change in the future ,you will be back to revamp and rewrite the strategy.

I wouldn't put any real money on this one strategy.sorry been there.
 
Top