Mr. Davison wrote in part, in his now classical form - I am one who holds the belief that IRVing is the best possible method for single seat elections, fear not, no one is going to eat you[r] first born. ---- D- Again the clueless folks who might have their first born eatten by H or S below ---- a really bad example of IRVing ---
H Hitler, S Stalin, W Washington 34 HWS 33 SWH 16 WSH 16 WHS 99 With IRV W loses. H beats S 50 to 49. Civil War II ??? W is the Condorcet Winner. If the first two choices are deemed desired/compromise (i.e. YES votes and/or Approval type votes), then H 50 S 49 W 99 (merely deemed tolerable by ALL the voters). How often will 2 extremists be the 2 highest choices with 3 choices remaining if IRV were to be used in real elections for a high office such as the chief executive officer of a nation ??? It only takes such an event one (repeat-- one) time before a *total* political disaster might happen (with perhaps the likes of Hitler or Stalin being elected with a false IRV majority). There are at least 2 types of data ---- absolute (such as YES/NO, 100 to 0 percent, etc.) or relative (1, 2, etc.) IRV is totally defective since it only uses a small part of the total relative data. Plurality is totally defective since it *really* only uses a small part of the total data. Thus the various methods can be ranked depending on how much of the total available data is used in the method. Condorcet Approval IRV Plurality
