It's clearer to express the numbers killed in proportion to the sizes of the group the victims belong to. There are many more women and men than trans people, of course, so giving just the counts doesn't tell us which group is more at risk of being killed.
But when you do the work (look at overall rate of people being killed, find out the size of the transgender population, and calculate a similar rate for them) you do find, indeed, in all the countries for which I have seen this data to exist in North America and Europe, that transgender people are at lower risk of being killed than the population in general.
The arguments trans activists use come from South America, especially Brazil, but for those arguments to make sense the trans rates of being killed there must be compared to the overall rates of homicide in the same countries. Homicides are common there, so the rate is higher for all people, too. And very high for sex workers, and transwomen are more likely to work in prostitution there.
A completely different kettle of fish is the assertion that the homicides of transgender people are always assumed to be hate crimes. Doubtless those exist and are dreadful and should be prevented (as should all killing), but most trans victims of homicide seem to be killed for similar 'reasons' as all victims, overall.
But on balance the data we have suggests that transgender people, as a group, are quite safe from homicide. And this should be a good thing, though we should always try to reduce homicide rates as much as possible for them as well as others.