Dacamic

There is the possibility this thread is more about promotion than enlightenment. On the chance my skepticism is unwarranted, I offer a link to the ROI calculation per Investopedia: http://www.investopedia.com/terms/r/returnoninvestment.asp. With respect, as Sevensa has said, SCAR is mathematically identical to ROI. Furthermore, the suggested interpretation for SCAR is to describe 1% x $20,000 as 2% x $10,000, which doesn't seem a revolutionary idea, either.

Did your tests include commissions and slippage?

As a possibly ironic twist to this thread, I suggest as a general rule being cautious about trading someone else's system. Certainly, we can appropriately borrow individual ideas from other people. In many instances, though, the final crafted product should reflect the aptitudes and preferences of the person trading it.

I'm curious about this, too.

Sevensa is correct. SCAR Ratio is a re-labeled and re-packaged simple return calculation (a rose by any other name). In answer to your earlier question, I constantly use return calculations. Yes, return will increase as performance improves during any given period of time.

The S&P 500 has rarely moved 50% higher year-to-year and never more than that. Based on that alone, a 100% up move in the DJIA is highly improbable.

Indeed, and it's displayed on a wall in my office.

Ah, yes ... re-opening old emotional scars by taking this trip down memory lane will be a great way to start my day ... I track my biggest losses by writing them down on an airplane "motion discomfort" bag tacked to my office wall. One stock appears several times on that memorial: Montana Power Company (MTP). To make a long story short, I suffered more than a few cuts trying to catch a following knife while this stock's price fell from $40 to $0 . It was inconceivable to me -- for a few reasons -- that this company could collapse so quickly. My equity has been hit by larger single-day losses, yet this string of "oopsies" is still notable.

You offer very good advice in your earlier post today. I wonder, though, if the approach you suggested is more attractive to people trading multiple securities than those trading only one. You and Hlm have mentioned reasons why "home runs" can be problematic, yet they seem more necessary -- or at least more tempting -- to single-security traders.

Daedalus ... Is it correct to assume you trade more than one security?

It's been awhile since I read Marty Schwartz's book (Pit Bull), and so may not correctly remember certain details. If my somewhat faded memories are reasonably accurate, Marty earned 25% compounded monthly returns using -- among other tools -- moving averages, stochastics and Terry Laundry's Magic T indicator for his trades. Discussions of this nature sometimes get hung up on definitional disputes about "system" and "indicator". As a guess, it's the term "mechanical" that is causing some people to have mental heartburn. To offer one perspective in that regard, my trading did not become consistently profitable until I began developing and trading mechanical systems, possibly because it afforded me a more structured environment.

An excellent point. The right choice for many people is to take their surfboards outta the water when there are 50-foot waves (which is what I did).

Scaling out (and scaling in) does have an intuitive appeal, and it is a commonly-used technique. But ... When scaling, expectancy becomes "spread" over multiple contracts. If we assume scaling out using three contracts, it seems reasonable to believe the first contract "scaled out" will have a lower expectancy than the "all-in" trades shown in Scenario A and Scenario B. By itself, the second contract isn't likely to have a sufficiently high expectancy to offset the first contract's shortfall; thus, it seems we would need the third contract to hit home runs (or at least triples) to match or exceed the expectancy of "all-in". Just a theory.

I agree and tend to focus more on expectancy.

When done properly, back tests can be very useful for developing trading systems. As you said, though, consideration must be given to doing more than simply running strategies against a batch of data. You mentioned three such additional factors maybe worth pondering: Monte Carlo simulations; parameter optimization; and, money/risk management. In regard to the original questions, my individual strategies are initially evaluated against at least 250 bars of data (I am an EOD trader).