Strategies over the long haul

[nelsonf]

My buddy and I have been busy coding for a few months and have noticed that some of our best efforts for 2010 don't play well in 09,08 etc. I've taken to optimizing over three years now. If we're getting consistent, strong results over three years (and give up some profitablilty in 2010), is this the preferable route to backtest? I guess we're trying to figure out how many different types of market conditions exist. Are we better just optimizing over the prior year and just keep tweaking as the date range we've used moves forward. Maybe optimize again after every month going forward?

Many Thanks,

Frank

[NinjaTrader_RyanM1]

Hi Frank,

Thank you for your post. There are no rules available for the best way of interpreting backtest results. It's a useful tool and good starting point before real time testing. Hopefully other community members can share their trading experience and provide some approaches for you. If you're looking to constantly reevaluate your parameters, a walk forward test is good for this:

http://www.ninjatrader.com/support/helpGuides/nt7/walk_forward_optimize_a_strate.htm