No your not.You "they" are getting their knowledge from circular reasoning,based on the assumption that our genes were identical to chimps.You cant hide half of the problem in a hypothetical ancestor.
I may be wrong,but that sounds like what you are doing.
I think you are misunderstanding this. It has nothing to do with "hiding half the problem in a hypothetical ancestor". It's more like basis statistics.
If you have two diverging populations that share a common ancestor, assuming the rate of divergence is about equal in both lineages (i.e. the mutation rate is the same), the net divergence between both populations will be approximately double the divergence in each population from the ancentral population. IOW, lineage A will accumulate a certain number of mutations and lineage B will also accumulate a certain number of mutations. If the mutations are strictly random, the overall divergence will be the mutations in A + the mutations in B, that make up the overall difference between A and B.*
The only assumption in this is the rate of divergence in each lineage, not that they shared an ancestor. What we are trying to see (which I've already shown repeatedly) is that relative rates of mutation are enough to account for those differences if they did share a common ancestor.
* Obviously there will likely be some overlap between populations, which is why the overall differences are logarithmic based on genome size vs time. But over a 5 million year period, it's pretty close to linear.