You’re not defying thermodynamics
.... but I do think a couple things are getting mixed up here.
If you’ve been sitting around 100kg for a couple months eating what you estimate is 2,000–2,500 calories, that’s actually the most useful data point you have. Whatever calculators or AI are telling you, your real-world maintenance is very likely somewhere in that range right now. The 3,300 number is almost certainly an overestimate, especially with activity factored in.
What likely happened is pretty simple: the intake that gave you a deficit at 120kg is now your maintenance at 100kg. As body weight drops, energy needs drop with it, and on top of that most people unconsciously move a bit less over time (even if they’re still training consistently).
“Crashing your metabolism” gets thrown around a lot. Metabolic adaptation is real, but it’s not large enough to turn a meaningful deficit into maintenance. It can make things slower and more frustrating, but it doesn’t override physics.
On the flip side, increasing calories doesn’t directly cause fat loss. The time that seems to work is when someone was really under eating.... to the point that their energy, movement, or training suffered, and bringing calories up improves those things. But that doesn’t sound like your situation, you said your energy and training are solid.
The biggest flag in your post is ballpark eating. At this stage, small inaccuracies matter a lot more than they did earlier on. A couple hundred calories either way is the difference between a deficit and maintenance.
If it were me, I’d:
tighten up tracking for 2–3 weeks (no estimating)
aim for a consistent 400–600 calorie deficit from your actual maintenance
keep protein where it is (you’re doing great there)
keep an eye on steps/overall movement, since that tends to drift down over time
You’ve already done the hard part dropping 20kg. This next phase is just less forgiving and requires a bit more precision, not a metabolism reset or a big calorie increase.
