It's not racist, it's just flawed.
The strongest predictors of crime are age, sex, and socio-economic status, i.e., young men from deprived backgrounds and living in poverty are statistically more likely to commit offences.
So when you compare a migrant population, which is usually made up of young men from poorer regions, against the entire native population, the figures are going to be skewed. Once you control for factors such as age, sex, and socio-economic status, much of the difference in offending rates is reduced and often becomes statistically insignificant.
These stats also often rely on arrest data rather than convictions or unique offenders. That's also an issue, because the same individual can be arrested multiple times, inflating the figures, especially within smaller populations.
But of course, that wouldn't generate outrage or keep people divided, so we just don't speak about it.
I mean the report that everyone quotes from the centre for migration control has literally been proven to be based on false data. The guy who wrote it deleted tabs, made basic arithmetic errors and changed numbers so that the data matched his chosen narrative, yet no one cares!