“I don't think I am? I am nodelocking and mixing the entire range, not just 1 hand.”

I must have misunderstood - so you aren’t using the Playing Against the Solution feature. Instead you’re using the node locking feature to fill in your estimated strategy for the flop, running the calculation, and noting the EV value. Then comparing this EV value to a sim that isn’t node locked.

This seems more useful, however, I sense that you may be comparing apples to oranges here at the end of the day. When you run the nodelocked sim, the opponent’s strategy will adapt to this strategy. Perhaps you’re better off exporting the frequency of each combos into Excel, and seeing how far you’re off, and address those that are off an unreasonable amount.

Hahah that is almost exactly what I am doing! Not each combo though, but I am comparing my frequency vs the optimal "GTO" frequency, and I am comparing my nodelock with the optimal sim and see where I went wrong (In what area's instead of single combo's).

I might add from myself that learning GTO depends on the stakes. I play 10NL and after learning some GTO my winrate dropped. After analysis I came to conclusion that as rfi I check ~38% on F (I check with many TP/M or L kicker and MP/LowP etc...). On the other hand, at these stakes betting for less than half a pot with a frequency close to 70% is more profitable.

Sounds like you've already done this but I'm reading Maria Konnikova's book and even though I'm two chapters in it's interesting how Eric Sleidel is already telling her that poker is not a game where you can just take 3 or 4 inputs, use GTO, and come up with a decision.

Every hand is unique and needs to be evaluated independently.