It would be usefull, if you would share your username on PokerStars and also make sure, you have opted in to
sharkscope, so we can check your actual playing history and results. As others have said graphs like this are really only usefull for cash games and maybe for SnGs, but not at all for large field MTTs, where most of the money is won or lost in a few hands at final tables.
Your total sample is 178k hands, which is probably less than 2.000 MTTs assuming an average of 100 hands played in each. And the simple fact is, that there is massive variance over such a sample, especially if you play during peak hours, where fields on PokerStars are large. For instance with a field size of 1.000 players, 150 places paying and a ROI of 10%, there is 34% risk of losing after 2.000 games.
Poker Tournament Variance Simulator – calculates variance for poker tournaments, MTTs and SNGs.
www.primedope.com
So most likely the main issue here is, that you dont understand the magnitude of variance in the game format, you have chosen to play. You feel, that your efforts "should" be rewarded by now, but the truth is, you have still not played enough to overcome variance. If you find this frustrating, then the solution is to change to a lower variance game. On PokerStars the best options for
bankroll building are regular cash games (not Zoom) or On Demand SnGs with up to 18 or 45 players.
Finally its seems like, you are very focused on the difference between your actual results (in chips won) and the all-in adjusted results. This is a trap, people often fall into, and then they feel, they have been unlucky, if they run below all-in adjusted EV. But the result of all-in before the river pots is only part of variance. So its entirely possible to run below all-in adjusted EV and still have been lucky overall.
Maybe you ran AA into KK more often than the opposite, maybe you completed your draws more often, than you should, maybe you flopped to many sets and so on and so forth. Also we cant change, how we run, when we are all-in with AK vs. JJ, so honestly its best to completely forget this all-in adjusted EV number especially in tournaments.
In cash games all-in adjusted EV can be a tad more usefull as a slightly less inaccurate metric for your actual performance. Plus on PokerStars you can choose to cash out and eliminate this part of variance entirely at a fee. But of course thats not possible in tournaments, because the goal of a tournament is to bust everyone, until a winner is found.