Recently I found a TAS video of 5-2 being completed with a 449 on the in-game-time, about five mario seconds faster than the optimal 5-3 method. Here it is:

If there technically is a faster way to complete 5-2 than 5-3, why is no one going for it? I don't know if I'm underestimating the difficulty in the helicopter trick, but I really do think this could be the future of insane runs.

I have been thinking that 5-2 is faster because I (An amateur runner) can get the same IGT as optimal 5-3 and I'm sure top-level runners can do much better than me.

I totally agree with you. I am also new to speedrunning, yet even with none of the speedups and bad movement I was still able to get 445, which ties the record for 5-3 with the propeller.

If any top-level runner sees this forum I recommend to them to try out this method!

the only reason I could see it being slower is because there are two pipe entries in 5-2 compared to just one in 5-3.

Even with a 449 it is slower because of the extra pipe transition losing more time than 4IGT (which is the difference between optimal 5-3 and optimal 5-2) even without loads since w/o loads only removes the black screen and not mario entering the pipe and the fade to black in and out.
Good idea tho, loses less for 24 runs because the extra complete black screen is removed.

Wow, I just spent hours getting good at this level. The pipe entries slipped my mind, sorry if I was a bit over excited. Anyway, back to the drawing board.

Just record an IL run for 5-2. Then your practice won’t be wasted.

as well, to be fair, a TAS obviusly is a perfect game of the level, and it wouldnt realy save as much time if a human isnt as perfect playing
this IL in the TAS presents some precise movements in getting the fastest slope-wiggler jumps and even mantaining them with a prop at the end of the level