Ignoramus, it would be better if you didn't challenge people but vote for them. Firstly, if I am not wrong challenges can be denied. Whilst it makes people seem a tad suspicious, valuing their life too much, defense of their persons right to live should be fine, to a certain extent. Voting is more certain, the lynch is guaranteed (i guess) if the person receives the most votes, challenges however, are only a 25% chance of being selected due to randomness. Furthermore, if we knew someone was likely scum, we would vote for him, not send a high battle rating samurai to go and duel him.

About the topic of battle ratings, I believe that it would be unlikely for Andres to not balance out all of these lovely goodies with Ninjas/Traitors with high battle ratings. After all, it would be a mockery if all of them had low battle ratings, they would be scared to even try a challenge attempt. These scum with high battle ratings are also unlikely to be revealed upon duel won. The duel of Glenn Vs Northnovas shows that duel results are rather ambiguous, it is this ambiguity that would allow the scum with high battle ratings to strike with impunity, albeit not rampantly. They would likely go around claiming the other person a tad suspicious and in the name of honor, the other samurai would accept to clear his name. If you take a look at Reenk's settlement game, upon a duel, the roles of the respective 2 would be revealed. Eg: Farmer swing hoe at priest, both unskilled, a mockery of the fine arts blah blah etc. Here, the write ups for the duels simply say that: Player X won Player Z. Perhaps the following duel would confirm for us if Andres simply writes duels as 'close ones' with action for the sake of story fodder , or whether it really was a close duel. I would suppose if the latter option is true, Andres might flip a dice and choose the winner if the battle ratings of the 2 people are equal I would guess.

As such, I would not make more evidence searches on duels until the next one occurs, for the first duel has yet to prove anything, it has but set a benchmark. We would do well to see the 'dominance' of 1 player versus another in the next write up for duels, or the same 'close game' style. It would likely take a few duels to accurately gauge...