The muddy battlefield hypothesis most generally states that tournaments that meaningfully distinguish between two teams of a given skill level do not meaningfully distinguish between teams of significantly higher or lower skill levels.
Ryan Westbrook first proposed the muddy battlefield hypothesis in regards to the 2006 Chicago Open History Doubles. The term returned in his critique of 2009 Penn Bowl, in which he claimed that length- and difficulty-controlled tournaments cannot simultaneously distinguish between two of the best teams in the tournament and two average teams.
In a tangent to 2009 ACF Fall discussion, Matt Weiner expanded on the idea by noting that tournaments that pack tossups with many difficult clues are often fine for the best teams but do not optimally distinguish between median and below teams. Furthermore, tournaments that use overly difficult answer lines are so difficult that they cannot meaningfully distinguish between any of the teams in the tournament.