Norgate Data and RealTest: Which Plan You Actually Need

Norgate Data is an end-of-day market data subscription for US, Australian and Canadian stocks, plus selected world futures and forex rates. It installs a local database on a Windows machine, updates it once a day through an application called Norgate Data Updater, and plugs directly into RealTest as a native data source.

People who pair it with RealTest are almost never buying it for price history, which is available free in several places. They are buying two things that free data does not have, which are companies that no longer exist and a record of which stocks were in which index on which date.

Why that matters more than the price history

Backtest a strategy on today's S&P 500 going back to 2000 and you have tested it on a list of survivors. Every company that went bankrupt, got acquired or fell out of the index is missing, and those are precisely the names that would have hurt. The result looks better than reality by a margin that grows with how far back you test and how small the companies are. We put numbers on that effect in the five backtesting mistakes that fake an edge.

Norgate solves this by keeping delisted securities in the database and by storing index membership as history rather than as a current list. In RealTest that becomes a variable you can test against. InSPX is true for a stock on the dates it was actually in the S&P 500, not the dates you happen to be running the test.

The part most people get wrong

Norgate sells its stock market data in tiers, and the two cheaper tiers cannot do survivorship-free backtesting at all. This is the most expensive mistake available in this decision, because nothing visibly breaks. The data imports, the backtest runs, and the number that comes out is wrong without saying so.

US stock packages as of August 2026, in USD.

Platinum is the entry point for the thing most systematic traders think they are buying. Silver and Gold give you currently listed securities and current index constituents, which means a test on the S&P 500 in either tier is a test on survivors by construction. No setting in RealTest can repair that, because the missing companies were never in the file.

Australian and Canadian packages follow the same structure at the same headline numbers in their own currencies, with Platinum reaching back to 1992 for the ASX and 1990 for Canada. A US OTC add-on is $49.50 for six months or $90 for twelve. Current figures are on the Norgate packages page.

How the integration works

RealTest reads Norgate through an Import section, which pulls the symbols you name into a single RealTest data file with an .rtd extension. You run the import once, then every backtest reads that file rather than touching Norgate again.

Import:
	DataSource:	Norgate
	IncludeList:	.S&P 500 Current & Past
	IncludeList:	.Russell 1000 Current & Past
	IncludeList:	SPY, QQQ
	StartDate:	Earliest
	EndDate:	Latest
	SaveAs:	universe.rtd

The words Current & Past are the whole point. Those are Norgate watchlists that resolve to every company that has ever been in the index over the period you hold data for, not the ones in it this morning. Drop those two words and you have rebuilt survivorship bias by hand, on a Platinum subscription you paid for.

Importing an index list also switches on membership variables inside your formulas.

Data:
	Setup:	InSPX and C < MA(C, 200) and RSI(2) < 10

InSPX, InRUI, InRUT, InNDX, InOEX and InDJIA cover the common US indexes, with InXAO and others for Australia and Canada. The authoritative list of which variable belongs to which index lives in constituency.csv in your RealTest install folder, usually C:\RealTest. Each row maps an index symbol to its boolean name, so when you need an index you have not used before, that file answers it faster than the documentation.

One detail saves a lot of time later. A membership variable only exists if that index's membership was loaded during import. Importing the Russell 3000 gives you InRUA and nothing else, so filtering that universe down to Russell 2000 names will silently fail. The fix is a Constituency line, which loads membership history without importing those constituents as tradable symbols.

Import:
	DataSource:	Norgate
	IncludeList:	.Russell 3000 Current & Past
	Constituency:	$RUT
	SaveAs:	r3000.rtd

Delisting itself is also readable. InfoDelist returns the date a symbol stopped trading, which is useful when you want to measure what the failures cost your strategy rather than assume it.

What Norgate does not do

It is end-of-day data. The company states plainly that it does not provide live quotes, delayed quotes, intraday or tick data. Any strategy that needs bars shorter than a day needs a different provider, and RealTest will happily test intraday bars from other sources.

The database is local and Windows-based, updated by running Norgate Data Updater. There is no cloud API in the sense of a hosted endpoint you query from anywhere, though Norgate does publish Python and Zipline packages that read the local database, alongside plugins for AmiBroker, Wealth-Lab and RightEdge.

What the whole setup costs

For a US equities systematic trader who wants results they can trust, the realistic first-year bill is a RealTest licence at $389 plus Norgate Platinum at $630, which is a little over $1,000. In year two it drops to the $159 RealTest extension plus the Norgate renewal, and the extension is optional since letting it lapse stops updates rather than the software.

Whether that is expensive depends on the alternative. The comparison is not against free data, it is against the cost of trading a strategy whose backtest was inflated by a few percent a year for reasons you could not see. One avoided year of trading a strategy that never had an edge covers the subscription several times.

How to set it up

Install Norgate Data Updater first and let it complete a full download before opening RealTest, since a partial database produces import errors that look like script errors. Then write a small import script with two or three ETFs and run it in RealTest with the -import flag rather than starting with a full index. When that produces an .rtd file, add the index watchlists you actually need.

Keep the import script separate from your strategy scripts. Imports take minutes and strategies take seconds, and mixing them means waiting for a data refresh every time you want to change an exit rule. One import file that writes several .rtd files, run weekly, is the arrangement most people settle on.

Once data is in place, the next question is what to test with it. Our strategies are built on this exact setup, with the index membership filters and delisted names included, so the backtest reports show what the rules did on the market as it was rather than on the survivors. They are in the RealTest strategy collection.