Not necessarily - there's an issue with desktop composition which affects tests (in my experience, Refresh Rate Multitool) where if it is enabled, it will skip frames, while in full-screen applications it does not. Hence why I prefer the RRM at full screen, as that takes care of any of these little foibles.
Frame skips in game will have a different nature - instead of an entire frame disappearing, part of one frame will render, then the frame buffer will update and the rest of the frame will render as the second frame. So, for example, at 90hz refresh, you will see the following:
0ms: Frame 1 is rendered and put into frame buffer. Monitor reads the framebuffer, and draws frame 1, starting from the top.
11.11ms: Frame 2 is rendered and placed into the frame buffer. The monitor has finished drawing 67% of frame 1. It then continues down the screen, drawing 33% of frame 2.
16.67ms: Frame 2 is still in the frame buffer. The monitor finishes its first full refresh cycle, drawing 2/3 of frame 1 and 1/3 of frame 2. It continues to draw frame 2 (which is still in the frame buffer) from the top.
22.22ms: Frame 3 is rendered and placed into the frame buffer. The monitor has finished drawing 33% of frame 2. It then continues down the screen, reading and drawing frame 3 for the bottom 67% of the screen.
33.33ms: Frame 4 is rendered and placed into the frame buffer. The monitor finishes drawing frame 3, and reads frame 4 from the buffer, and draws frame 4.
In this cycle, we drew the top 2/3 of frame 1, the top 1/3 and bottom 1/3 of frame 2, and the bottom 2/3 of frame 3. We never see a single, full frame drawn, but no frame is "skipped".
So in games, instead of skipping entire frames as seen in RRM and the test by Blurbusters (which are purposely designed in a way that apparently skips entire frames if frame-skipping is detected), you should notice an increase in screen-tearing.