Help with finding proper move sequence of an old OTB game

Hello all, I'm not 100% sure this would be the correct place to post this, but it is a game and I request analysis. I have the notations from of a classical OTB game I played from years back that I had fond memories of winning because the guy was rated more than 300 points higher than me at the time and would regularly slap me during club practice. The only problem is I think I completely forgot to record a few moves and to add to the confusion, probably wrote down nonsense during crucial moments in the middle game.

Anyways, here's a picture of the original notation.

I'm pretty sure it is accurate up to 16. aRb1 e5 which I've put into this study. lichess.org/study/56N4zZ7E

The garbage and missing notation starts from there, but I do remember 18. Bd3 Rd8 19. Qxh7+ Kf8 happened at some point in the game. For the life of me I cannot figure out what possible combinations of moves would lead to my queen being able to get to h7 from the position indicated in the study and any ideas would be greatly appreciated.

Almost as scary as a Fischer scoresheet.

"For the life of me I cannot figure out what possible combinations of moves would lead to my queen being able to get to h7 from the position indicated in the study"

I too am stumped.

@SwampTiger
I used to have poor handwriting but not that bad. Also, if the power of my lenses increase, it would be because of you!

Well, let's come to the game. Move 17 seems Qf5 gxf5. Wait, a Queen sacrifice? And if Queen is captured, how can it give check in next 2 moves? Wait, Qf5 is illegal since e4 square has White pawn!
So, with respect to possible moves there and what you wrote, it seems f5.
And if next move is Bd3 and Rd8 , then Qxh7+ is legal since three White pieces are in between the diagonal.
Please can anyone know what's there, because if I pay more attention to this, I might go in coma!

I ask God why this man saved the 'improper notations'!

Actually, this all has certain intriguing resemblances to retrograde analysis. :)

Anyway, it seems like Black must've played f5 in there (in order to get the diagonal leading to h7 opened up). Problem though is that he has to play Kh8 first (to save his queen from the discovery)...and then Qxh7 will be mate (not check).

Hey OP, how about this: can you remember what the final position was (and work your way back)?

@Akbar2thegreat Yes, I think f5 must've been played, and I saved these notations because they were in a 100 game notebook I had as a kid haha.

@MrPushwood I remember getting the queen in front of the bishop to make a battery, him moving the rook to give an escape square for his king and playing Qxh7, from there on out it was mostly cleanup.

I've tried for a while to recreate a possibility in light of the suggestions, and I've updated the study lichess.org/study/56N4zZ7E. However later on, I do remember he did get his bishop to take on h2+ to delay the inevitable and he also blocked using his knight on f8 which isn't compatible with this variation but I think I've spent enough time trying to figure this out lol.

I suppose the greatest victory of my schoolboy days will have to be forever shrouded in mystery.

"I suppose the greatest victory of my schoolboy days will have to be forever shrouded in mystery."
Maybe it's better that way, you might be shocked by the quality of the moves...

@SwampTiger
The moves that you made in study compared with the notation picture aren't fitting well and some major part is missing.
And I don't think, we can get over it.
Probably, this problem can't be solved and makes record for the greatest unsolved puzzle in history!

OR: ask Magnus Carlsen, he'll know (jk)