Radio waves go faster (speed of light) than through a medium (copper).

Except that copper ethernet is baseband, so it's not radio waves. WiFi is still faster than copper AFAIK (there was a huge debate about this between youtubers not that long ago), at least for signalling, but the difference is smaller than you think. light (which is EM, the same waves as radio/WiFi) through glass is about 2/3rds c (aka the speed of light), and it's actually a lot slower than ethernet or WiFi for propagation delay, however, WiFi must use CSMA/CA as well as other tricks to ensure it doesn't step on itself, and that it doesn't step on other sources of radio interference (Microwave ovens, wireless controllers (like xbox), bluetooth, zigbee, etc, on 2.4Ghz and stuff like RADAR on 5Ghz). It's half-duplex, so only one station can transmit at a time, hense CSMA/CA being required, where ethernet doesn't need any collision avoidance or detection except for rare cases of 10/100 half duplex, all gigabit is full duplex. Half duplex on wireline networks is basically eliminated at this point, so it's little more than a footnote.

Factoring all this in, getting the signal down the line, WiFi loses in almost every case, due to all the considerations it needs to take into account.

There are lots of factors that can cause jitter on WiFi, and it's mostly outside of your control if you're living somewhere more densely populated. My apartment randomly gets a lot of noise, and as a result my WiFi starts to get unacceptable amounts of packet loss and jitter. It doesn't happen often enough to motivate the effort for me to go around signal analyzing, but still...

Wi-Fi has constant retransmissions. This adds perceptible latency because the checksum check, turnaround, and packet transmission add a lot of time compared to the speed of light through air across 3 meters.