Low rolls for extended period of time

I understand how the quantum roll system works and I know about the chart you can look at to see the distribution. However the fact still remains that I have been getting really bad rolls for quite some time. I typically don't get above a 10, in fact the majority of the time I seem to roll around 5 or lower. I do get the occasional decent result; but more often the not around 5 or lower is the norm for me. I have done some experimentation regarding this, and it would seem that when I make a single die roll with a few minutes in between its a low result. But if I make a roll with about 3+ I get a better spread, still not great most of the time mind you but better. When I increase this to about 5+ I actually get a good spread, I must note that in this manner of rolling multiple dice all at once the total added together is still only about average most of the time. Where this start to get interesting though is when I make a lot of single rolls over a relatively short period of time (usually when testing macros). I start to see what I would call a close to perfect spread, a wide variety of different results ranging from min. to max. Now my gm seems to have the same problem in reverse, they consistently get good rolls and very rarely get bad ones. Now I don't understand how the information regarding rolls is compiled into that nifty little chart you have. But is it possible that while it appears the system is truly random and every number is coming up in a relatively consistent amount between all values. Is it possible that your system isn't taking into account who is rolling those numbers. What I mean by that is, is it possible that most everyone sees a fairly random series of rolls but some users seem to see specific numbers pop up more often than others? And I'm not talking just bad rolls Here, but is it possible that some users see one or two specific number(s) far more often than others? At this point it's the only thing I have left to explain how my rolls are far more often than not around 5 or less (again not all the time but at least enough to be anomalous). This is the last thing I can think of doing before just making a new account in the hopes that it miraculously fixes my problem.

Andreas has put together a good response to this very common question/complaint. You could also try the following a few times in chat. You should get something in the ballpark of 10.5 [[(1000d20)/1000]]

The die roller has no idea who is rolling or what it has given them before. If your roll is number 1,000,000 of the people who have rolled again, and a thousand people roll between your first and second commands, that will be roll number 1,001,000 for that hour. Random doesn't mean consistently even distribution. You are likely just sitting on an end of a distribution curve, and maybe some confirmation bias is at play.

There is an api floating around that is very new called RollStats that will install and record the rolls you make and spit out a leaderboard and lifetime/session results for you to see if you are one to care about the odds.  I, as the DM, have noticed the roller to be very accurate, however players do not roll nearly as much as the DM and it feels very broken at times.   Glug and Shayla were advantage rollers, Wigs was spamming his longsword key cause I installed 30 new longsword attack sounds and he was testing it out in the session.