Grinding it out, day after day

Lescor,

How would you approach backtesting a completely discretionary strategy based on 70% price action 20% fundamentals and 10% indicators (stoch, rsi, macd, bands, etc.)

I'm trying to develop an "edge" instead of relying on just skill/experience and money/risk mgmt.

Thanks a bunch.
 
If it's discretionary I don't see how you'd do anything other than going back and looking through charts one at a time. Even then you might have a bias because you know what happened.

That's why I'm a fan of forward testing in real time with real money, but with the smallest size possible until you start building some stats.
 
yes i recall looking at the tuco trading sheet and the #1 guy was like 380-400k for 2007 and that was dustin i believe. but why anyone cares what somebody else makes is beyone me.many guys have many accounts and will show there best accounts.
 
Quote from piggie2000:

yes i recall looking at the tuco trading sheet and the #1 guy was like 380-400k for 2007 and that was dustin i believe. but why anyone cares what somebody else makes is beyone me.many guys have many accounts and will show there best accounts.
More...

Where is Dustin?
 
Quote from lescor:

If it's discretionary I don't see how you'd do anything other than going back and looking through charts one at a time. Even then you might have a bias because you know what happened.

That's why I'm a fan of forward testing in real time with real money, but with the smallest size possible until you start building some stats.
More...

Thank you. I appreciate your advice as I'm sure many other traders here do too.

However, one of my main rules is to not lose money (i.e. "preserve capital" as you say). Therefore I prefer to papertrade new strategies or modifications to current strategies to test their performance without risk.

Although the psychological fear of losing real money isn't present in a papertrading account, at least the mechanical aspects of a new strategy can be worked on.
 
Quote from lescor:

That's why I'm a fan of forward testing in real time with real money, but with the smallest size possible until you start building some stats.
More...

Good advice. This is where I am now. I came up with an arsenal of setups/signals I really like and I still do my bar-by-bar backtesting, but am only trading live 1 lot (oil futures) until I can duplicate the results based on taking all trades, not cherry-picking trades. Once I reach the point of a) taking all valid trades, and b) determining that the results are very close to the (unintentionally biased) manual backtesting, I'll be ready to size up.
 
+5k for the week, another snoozer. Daily pl +3, -3, +3, 0, +2. Traded 500k shrs. Some good intraday moves to fade, made money each day with that strat, just not with enough size. Seemed to be one big loser most days to keep things from being really good. $1,000 brain fart today didn't help either. Swing trades seem to have turned a corner with some tweaking, gradually eking out profits there.

Away from trading I'm enjoying my summer, but the trading itself has been largely frustrating.
 
You must log in or register to reply here.
Back
Top