True? Well, almost. As a general rule the 50% figure can be applied but the real answer as to how much rope strength is lost is "it depends".
Have a look at this study conducted by Dave Richards, Technical Director with the Cordage Institute in the US. His team tested three different ropes types (and diameter) using eight different knot combinations common to the industry. The results show how each knot performed relative to the break strength of each particular rope - makes very interesting reading (and the graphs make comparisons easy).One common theme throughout the testing was the importance of tying the knot carefully, paying attention to detail. Obviously a knot tied in a sloppy manner will not perform as expected. Inappropriate knot tying is one of the reasons the industry is moving towards certified splices in climbing systems. According to Thilo Beeker, the International Representative at the 2009 NZ TCC we are going to see more and more pressure being applied to have the number of knots in the climbing system reduced and all splices certified.
Of course you could argue knot tying is part of the art of tree climbing (and rigging). If you can't tie a descent knot you shouldn't be climbing a tree in the first place! That said, given that even the best tied, dressed and set knot will reduce the strength of the rope (by varying percentages) there are plenty of situations where a splice would be more appropriate.
Tachyon tested by DM Standens in Auckland broke at the knot.