Wikipedia states that:
Machiavellianism has been found to be negatively correlated with Agreeableness and Conscientiousness
Next, agreeableness article has following words:
The lower-level traits, or facets, grouped under agreeableness are: trust, straightforwardness, altruism, compliance, modesty, and tender-mindedness.
And article on Machiavellianism in some way is contradictory:
It was found that in a trust game, those with high MACH-IV scores tended to follow homo economicus' equilibrium strategies while those with low MACH-IV scores tended to deviate from the equilibrium
In simple words that means that high machs are closer to optimal strategy in trust games than low machs. If they are mistrustful, why are they better in trust games?
That's not the only contradiction. Let's look at conscientiousness article:
People who score low on conscientiousness tend to be laid back, less goal-oriented, and less driven by success
And now again Machiavellianism article:
High Machs admitted to focusing on unmitigated achievement and winning at any cost.
But enough Wikipedia, there is more ambiguousity in this research article (that references a bunch of other articles):
There are at least two main paradigms of partner selection and both can be evidenced with previous research to be relevant to the partner selection of Machiavellians. On the one hand, similarity in level of affiliation, intimacy needs and personality traits has particular importance in ideal standards and mate choice. Previous studies provided countless evidence supporting the similarity hypothesis
Accordingly, it is also possible that a dominant high-Mach who demands attention can form the most stable – though dysfunctional and asymmetric – relationship with a submissive and permissive low-Mach partner who he or she can manipulate and exploit
Again ambiguousity. The only explanation I can propose is that there is no such thing as Machiavellianism. There are [at least] two types of personality which Machiavellianism tries to describe. Is there any concept that tried to resolve these issues? Obviuosly, Mach-IV can't separate these two types of personalities. So, are there any tests that can do it?
P.S. I expect that those "machiavellianists" who are good in trust games usually tend to choose similar [to themselves] partners for relationships and possibly, they are more goal-oriented (and should they really be a part of dark triad?). Also I won't be surprised if these two types of "machiavellianists" will differ greatly in emotional intelligence and mind theory.