I am not a day trader so I base my tests on end of day information

In the evening I fetch my data from the stocks I am interested in. This is all stocks on the Swedish stock exchange. I get them from a server, buying the data support from Börsinsikt. I do a lot on the incoming data before I decide what to trade and what not to trade. The decisions are based on my systems, all of them tested against all stocks on the Swedish stock exchange, all of them statistically profitable on the dates I have run the tests against. The assumption is always the same in the systems: I will buy the stock the next day, and I will buy it for whatever price is currently set by the market. I will then keep the stock until one evening I get a signal to sell it, and I will then sell it the following day, to whatever price is set by the market.

My test environment is Tradestation, running on data from Börsinsikt plus own special programs

I keep all old stock information, also for stocks nowadays not available on the market (if you had traded then, the profit or loss would have been there, independent of if the stock exists today or not). I enter the system in such a way that I can later on run another program on the trade information, doing calculations on cost for commission, profit/loss, win per trade, percentage winning trades, winning trades in a row, losing trades in a row, maximum drawback, largest losing trade, largest winning trade, total profit, number of trades, number of winning trades, number of losing trades, profit earned on winning trades, profit earned on losing trades.

The known limitations in Tradestation does not limit my tests