While I can't calculate the liklihood of breaking even over 2 very different games, here's my thoughts on, with a few assumptions that may be relevant or not, to losing your $300 (30 units) in 10 hours.
Should u choose to isolate your $3 BJ results, I can probably do that.
Assuming you played 300 hands in those 10 hours, since you were only playing positive running counts, with an average advantage of basically nothing, just my assumption, you are indeed outside your 90% range, say 93%, of losing that much or more, or around -1.45 standard deviations or so.
Even assuming a BS player playing 60 hands an hour, 600 hands in total, it would be around 83% of losing 30 units or more, less than 1 standard deviation.
So, my guess, somewhere in that range.
Hope that the liklihood that you have likley suffered nothing more than what seems like nothing more than a little bad luck helps you in some way.
As you know, 10 hours ain't crap anyway.
The more you can estimate how many you hands you played at what avg HA,
the more accurate we can be!
And, should you ever bet more than $10 on a hand, because you're pissed or whatever, it can make a big difference in these calculations. The main thing is, it ain't dollars, it's units.
Do you, or did you in those 10 hours, really never ever bet more than $10 come hell or high water? Just curious. If so, I congratulate you on your discipline