http://www.husng.com/content/given-m...true-winrate-0

according to the charts in the website above, after I play 3000 games with even a 52% win rate, I'll have just over 2% chance of breaking even or lose. and if I have a 54% win rate, I'll have less than 0.1% chance of breaking even or lose over 3000 games. but so many graphs I've seen on sharkscope from these players with 3.5% roi (idk what percentage win rate that is, 54% maybe) they have many many breakeven stretches over 1000, 2000, 3000 games.

like this graph for example, 2.68% ROI, and yet look at game 2591 to the end at 5564 its just breaking even.

I have been curious of late in trying to determine what the achievable win percentage is in the game or if that number is somehow limited by the structure of hypers.

For example, what could be achieved if we took a great high stakes Hyper pro like DC and dropped him down to the $30 level. Assuming he gave it his best effort, what would his win percentage be over 20,000 games? Is 60% even realistically possible? Yes, I know anything is possible; but, I rarely see hyper players over a 10,000 sample with a win percentage over 55%.

Sorry to derail OP, but as to your question I wonder if there is a difference between the first 3,000 or any stretch of 3,000. There should not be if your numbers are correct. Maybe that chart is simply an example of that 1% or .1% you referenced.

Just glanced at it, but is rake the issue? A 50% win rate is losing after rake, not break even.

Remember taking it like that is huge simplification.
Nobody said you have 2.68% in every single game. There might be stretches where you're closer to 0% or even negative, and streches where your ev roi, vs gives opponents in it, might be higher.

There's also a big difference between taking a 3k sample size from a population with a known win rate (like in the link you gave) and estimating a win rate based on a 3k sample (what you're doing).

In you're example, the player in question might be a break even player who ran hot for a small stretch at the start (or might have encountered some bad players during this stretch as suggested), which is still influencing his ROI by the time he hits game 5k.

Spot on