The 99% Guarantee

The 99% Guarantee

September 27, 2017

I’ve said it before and I’ll say it again: the number one thing that determines trading success is trust.

Trust your system? Anything is possible.

Don’t trust your system? Losing everything is possible.

So how do we get there? How do we learn to trust our system? There are two paths.

One is complete and utter belief in a timeless principle.

For example, we could trust the idea that the stock market goes up over time, no matter what. No matter the Great Depression or the Internet Bust of the 2000s or the Great Margin Call of 2008, we trust that the stock market will always go up.

That’s what Index Fund people believe–and they’re pretty adamant about it. Warren Buffett believes this without question. Jack Bogle swears by it. Tony Robbins wrote two books about it. And Mr. Money Mustache actually lives it (all of his income is based on the returns of an Index Fund).

These people will slap you in the face and tell you nothing else is better than believing in the market (and some of them will also tell you that nothing else works).

The stock market goes up over time and that’s the infallible truth.

Of course, it doesn’t go up all the time. In 2008, one of the most popular Index Funds, Vanguard’s VFINX, dropped 37.02%. And in the three-year period from 2007-2009, an Index Funder would have seen his account DECREASE by 5.14%. Three years of waiting and no income to show for it, only losses.

But for the people who trust this belief, that three-year losing period means nothing. It wouldn’t shake them out. They wouldn’t even bat an eye.

And they would’ve been rewarded for it with a 103% gain since 2009. That’s the power of trust.

But what if you don’t like three years with no gains? Or what if you think you can do better than the Index Fund average of 7.25% per year?

Then there’s another path.

This path means you’re a trader. Instead of just sitting back and taking your wins and losses as they come, a trader tries to find an edge and exploit it to significant gains. How do we find those gains?


We spend hours and hours testing possible edges until we find something we like, something that does better than an Index Fund.

So once we find an edge, what happens then? Do we go live in the market?

That depends. On what?


Do we trust our edge? How did we get it? Did we test too much and over-fit the data? Did we only find an edge that works in a specific market environment that will never occur again? Or did we find an edge that will persist over the long-term (like the rise of the stock market)?

How can we be confident our edge will hold up?

The answer for some people lies in the quality of the data itself.

If you’ve decided to base your financial future on the results of MT4 testing, for example, then there’s a problem. Is the data good data?

When we run a regular MT4 test on Strategy Tester and click on the Report tab to get our life-changing results, we see a little notation in the top right of our Report. It says, “Modeling quality” is “[some number].”

If the modeling quality number is low, say under 90%, then trust might become an issue because that means our data isn’t entirely accurate. If it’s not accurate, it’s easy to believe our research isn’t accurate.

Furthermore, after looking at modeling quality, we see a scary notation on the left. That notation says, “Mismatched chart errors.” If the modeling quality is low, then you’ll see a big number of mismatched chart errors. If there is a ton of chart errors noted, then trust can really become an issue.

Is there a cure?

Yes! There are places to find “perfect” data, data that has a 99.9% modeling quality and zero mismatched errors. I spent last weekend finding that data and loading it into my desktop computer. (Yes, it actually did take two full days. Not the most fun I’ve ever had.)

After my weekend slog, I finally was able to run my test going back to 2005 on my USDJPY Hornet robot and here’s a screenshot of what it looked like:

99.90% modeling quality on the left and 0 mismatched chart errors on the right. Now that’s some high quality data!

Considering that’s as good as it gets, what does that do for our trust? Does it get us to the level of belief that Index Funders have? Can we trade this live and never be upset when times get tough because we have 99.9% data to back us up?

Just out of curiosity, I ran the same test on my laptop’s MT4 using scraggly old data that comes standard with a free Oanda MT4 download.

This grubby data Report shows a modeling quality of “n/a” (whatever that means!) and 305,864 mismatched chart errors. Yikes! Who would trust that data?

Before I threw my laptop out the window, I thought I’d take a quick look at the numbers. How different were the two Reports?

Quick comparison (using a tiny 0.15 trade size from ’05-now):

  • 99% data: profit $4,402/max drawdown -$255.88/no losing years
  • “Bad” data: profit $4,971/max drawdown -$277.10/no losing years

Hmm. They’re about the same. The difference being the 99% data took less trades and, thus, made less money. Why would it take less trades? The indicator readings between the two platforms are slightly different.

So we’re back to square one. What do you trust?

If you trust your belief in the stock market, then you can get started today and make your 7-8% per year fairly reliably.

If you trust your belief in 99% testing, then you can do way better than an Index Fund with the same amount of drawdown.

If you think imperfect data is just fine (because it’s so similar to “perfect” data), then you can move forward and not spend a whole weekend waiting for data to load.

All of that is possible.

It’s just a matter of trust.