The SARS-CoV-2 early strains were pretty bad at transmitting. Most chains of infections actually died out. It was estimated that it took several introductions

The SARS-CoV-2 early strains were pretty bad at transmitting. Most chains of infections actually died out. It was estimated that it took several introductions of the virus before outbreaks actually started. It still mostly infected people you had close contact with over a sustained interval. People who were out jogging weren't actually infecting the people they were running past, it was people in the same household, or people who were eating dinner together. And it is still quite severe and lethal once it finds a few people who are over 50 with comorbidities or anyone over 70 (it doesn't take anything particularly special to kill you if you're over 70). If the researchers were all in their 20s and never had any cross generational contact then it could transmit undetected, but people mingle across generations more than the average college student or young adult working in a FAANG/startup (particularly in China). And if you hypothesize cryptic spread then you have to explain the geographic clustering of the initial cases around the wet market. Now you have a worse problem because the lab worker gave it to someone who gave it to someone who gave it to someone else who somehow spread it in the wet market to make it look exactly like that was a spillover event and all the other chains died out without putting anyone in the hospital.