Nothing dropped. The disparity in minutes played is because you used official numbers which is gold standard but I used ASA because ASA presents its data in a way that is most useful for manipulation and comparisons. But the most quirky thing about ASA is that it counts added time in players minute counts, so those totals never match official MLS figures. If you play a full game, then ASA credits you with anywhere from 93 to 99 minutes and maybe even higher. The upside is it better reflects actual minutes played, especially for late game subs. Officially if you enter a game at 89' and there were 4 added minutes, you played 1 minute. ASA credits you with 5. But it makes comparisons across different data sources more complicated.
Beyond that, I fully ignored both 2020 and 2021 in my analysis. I only based it on 2019, and for reasons I previously explained, I do not think he is likely to repeat that performance even without factoring in injury rehab risk. I'm playing percentages and likelihoods, so I could be wrong even if my analysis is correct, but I think it's quite likely that if he plays at least as much in 2022 as he did in 2019, his per 90 numbers will be lower. Plus I can't ignore how often he was hurt in 2019. Unless that changes it puts a strong cap on his potential. For me the best case for upside is the possibility that he will get fewer soft tissue knocks, which you sometimes (but not usually) see after a major rehab. Significantly more minutes could mean more goals. But I think 2019 was a career best type season and he's unlikely to score at that rate again. I hope I'm wrong.