Reply to Thread

Message:: [QUOTE="brg5658, post: 1949452, member: 29751"]The biggest problem is that there is no gold standard, so even the definition of "accurate" is not fixed.  Even two extremely experienced people supposedly using the same ANA standards will often disagree on a coin's numerical grade.  To measure accuracy you must have a known (true) fact to which to compare.  Coin grading is inherently subjective, so that's never the case. What you can measure is precision (or repeatability) of results.  Which is what the article you're referencing did.  And, in my opinion, precision (or consistency or repeatability of grading) is just as important as accuracy.  In the study you're referring to (I believe it was back in 2003), PCGS actually had the lowest levels of precision -- that is, they didn't grade the same exact coin the same thing upon multiple submissions.  Crack-out masters thrive on the imprecision of TPGs, not on their immeasurable accuracy.  If you play the crack-out game long enough and you're smart enough to know when to stop trying your luck, then we end up with a bunch of "low end for the grade" coins in holders that they shouldn't be in.  It's usually pretty clear when a coin has been "maxed out" and no person in their right mind would dare crack it out and resubmit.  It has been milked for all it's worth! <img src="styles/default/xenforo/clear.png" class="mceSmilieSprite mceSmilie2" alt=";)" unselectable="on" unselectable="on" />[/QUOTE]