They kinda have been at the end of Maxwell and during the Pascal era. GTX 980 in laptops was the first "desktop-class" mobile GPU, released with much fanfare from Nvidia as a different card from the previous GTX 980M. And it actually held very close to a regular 980:
https://www.notebookcheck.net/NVIDIA-GeForce-GTX-980-Notebook.150599.0.html
This was the reason why starting with 10xx series Nvidia dropped the "M" suffix in model names entirely. Performance parity with desktop became a big part of their marketing for a while. Except for Max-Q models, 10xx mobile chips where often within 5-15% of desktop counterparts in games. There were of course some outliers, but generally the performance delta was small - and notebooks used the same chips as desktops, only clocked lower and possibly better binned. You can check it on
https://www.notebookcheck.net/Computer-Games-on-Laptop-Graphics-Cards.13849.0.html - just search for GTX 1060/1070/1080, allow both desktop and laptop GPUs and see the results. This also mostly holds for non-Max-Q variants of mobile 16xx GPUs - the performance gap might be slightly higher there, but it's still pretty small.
Then came the Max-Qs, the 2xxx series, multiple TDP variants and while the names stayed the same, the performance of mobile GPUs started lagging behind. Nvidia also started using different chip configurations or even different chips in notebook variants altogether... And now the disparity is almost as big as before, only model names are more confusing.