No, it's pretty much on the ball.
BPM is a question of how'd you break it down on sheet, and there isn't necessarily one correct answer. In notation you could do 216 BPM with 3/4 (or 12/4 if we're to keep bars the same size) time signature instead of 72 BPM with triplets and hexaplets for same end effect.
Odd section.
Before the acceleration part I'd start with +10 ms offset on first timing point as it feels slightly early to me.
I think because of that early timing in beginning you held 145-146 too long while falling behind in sync, but then after delayed 146-147 you jumped straight to 150-151 too early to catch up. It didn't work because subsequent resets on 151 were not snapped to one another and reintroduced delay back again with late offsets. You don't need to reset every bar for the accelerating part if bpm doesn't change.
Last timing was fine, but I shifted it 1ms just to match cleanly with the transition.
Timing points (Insane)[TimingPoints]
4095,416.666666666667,4,1,0,50,1,0
4095,-125,4,1,0,50,0,0
15761,-133.333333333333,4,1,0,50,0,0
17428,-90.9090909090909,4,1,0,75,0,1
29095,-133.333333333333,4,1,0,60,0,1
30761,-133.333333333333,4,1,0,60,0,0
44095,-111.111111111111,4,1,0,70,0,0
64095,-99.9999999999999,4,1,0,80,0,0
70761,-83.3333333333333,4,1,0,80,0,1
97428,-133.333333333333,4,1,0,60,0,0
124095,-90.9090909090909,4,1,0,60,0,0
130761,-76.9230769230769,4,1,0,60,0,0
135331,-66.6666666666667,4,1,0,60,0,0
135345,-66.6666666666667,4,1,0,60,0,0
136386,-83.3333333333333,4,1,0,60,0,0
137428,-100,4,1,0,60,0,1
164095,-125,4,1,0,60,0,0
167428,410.958904109589,4,1,0,60,1,0
170715,408.163265306122,4,1,0,60,1,0
173980,405.405405405405,4,1,0,60,1,0
177223,402.684563758389,4,1,0,60,1,0
180444,400,4,1,0,60,1,0
183643,397.350993377483,4,1,0,60,1,0
185232,394.736842105263,4,1,0,60,1,0
189968,416.666666666667,4,1,0,60,1,0
190801,416.666666666667,4,1,0,60,1,0
190801,-200,4,1,0,60,0,0
204134,-100,4,1,0,50,0,1
217467,-100,4,1,0,50,0,0
It's going to need a resnap and might mess some sliders up.