The evidence suggests that the death of George Floyd both represents and is an instance of the oppression of black people in the United States, in the sense of the existence of both prejudice and disparate negative outcomes. This does not by any means rule out a possible effect of apparent wealth.
First, as noted in another answer, African-Americans are only around twice as likely to be below the poverty line as European-Americans. However, data from 2014 suggest that African-Americans are five times as likely to be killed by police as European-Americans (not the most unbiased source, but I have seen the data elsewhere). Although I have not tested for significance, this would seem to indicate that poverty alone cannot explain higher rates of police killings of black Americans. There is a strong component of racial bias as well. Some research has also found that the race of a hypothetical target can influence the likelihood that subjects will employ force in laboratory settings, with them being more likely to use force against black targets.
In addition, there are studies that have found a correlation between the levels of racism in a state and the rate of police shootings of unarmed victims.
Researchers then created an index of structural racism at the state
level, which they believe is the first of its kind. Looking at
measures of black-white residential segregation and disparities in
economic status employment status, educational attainment and
incarceration rates, the index was cross-referenced with data on
police shootings.
For every 10-point increase in the state racism index, researchers say
they saw a 24 percent increase in the ratio of police shootings of
unarmed victims.
Obviously, correlation is not causation, but this is telling. Overall, the evidence seems clear that to the extent that Floyd's death represents the disproportionate deaths of black Americans at the hands of police, it represents racial disparities, not just economic disparities.
Further, we should consider the individual case, not just statistics. The numbers mentioned previously are sufficiently striking that one might reasonably be inclined to suppose race played a role in a randomly selected shooting of an unarmed black victim (or even an armed black victim). In the specific case of George Floyd, the officer who killed him had a history of excessive use of force, particularly against black suspects. Several people interviewed who had known said that they viewed him as prejudiced against black people. For instance, the owner of a nightclub with a primarily Latino clientele said that he was comfortable with the usual crowd, but that when black clients would come in, he had a propensity to get violent. In short, the officer who killed Floyd seems to have been bigoted and inclined to act out violently against the targets of his bigotry.
Further, since race and poverty are correlated, in the hypothetical (and counterfactual) case in which police officers exhibited no racial bias but were instead only biased against low-income individuals, they would still kill black people at higher rates. Thus Floyd's death would still represent the oppression of black Americans, but at the economic level, as opposed to the level of police bias.
Finally, from an ethical point of view, if the police were strictly unconcerned with race but still perfectly willing to employ disproportionate force against targets whom they perceived as poor, that would not precisely be encouraging.