A strong argument for ethical fitnessism is that by definition natural selection will cause organisms to tend to act according to ethical fitnessism.
Err... no.
Organisms do not act according to ethical fitnessism -- you define fitnessism as whatever behaviour was picked by natural selection. Accordingly, there is no "strong argument", it's just the definition of your neologism.
Fitnessist behaviour will out-compete other behaviour
No, because if you're looking backwards in time, conditions change and what used to be adaptive might be counterproductive now. And if you're looking forward in time, you have to make guesses about what will be selected for in the future and I don't know why would your guesses be correct.
It is true that “organisms do not act according to ethical fitnessism”, but that is not what I stated. What is true is that organisms tend to act according to ethical fitnessism, which is what I stated. It is true by definition. I believe that a strong argument for a moral theory is that it is being practiced more than other moral theories.
As a consequentialist it is hard to predict which actions in fact will maximize the intrinsic value and in retrospect a behaviour that might have been seen as favourable at the time can have been a huge mistake in the long run and such behaviour will not be favoured by natural selection. Natural selection might seem short-sighted but it is not.
I noticed that there has been some earlier discussion about Sam Harris’s Moral Landscape Challenge here at LW. As a writer on the Swedish politico-philosophical blog The Inverted Fable of Reality, I would like to share a response to the challenge, written by our main contributor, which I believe is interesting to read even if you are not familiar with The Moral Landscape or its content. See this link for the response and a short explanation of the challenge.
The response takes a different approach to most responses to the challenge. It is divided into four parts and starts by asking which ethic is most compatible with science and reality and finally tries to answer this question.