In cvdata, when I turn on wonging it gives me numbers that don't seem correct. When I set back counting to start at shuffle point then stop in a TC less than -1 it gives me a positive win rate when the count is in the negative. Why does it do this? Are the numbers I get for win rate correct?