4x4 Yau is arguably the fastest and efficient method to solve a 4x4. The Yau method was proposed in 2009 by Robert Yau, and since then has been used by most of the top speedcubers worldwide. The method is a slight variation of the standard reduction method for 4x4, which surprisingly produces a much faster and even more efficient solves.
To begin with, what is the Yau method and why is it used by so many of the top speedcubers worldwide? Well, in layman terms, the Yau method is basically a reordering of the sequence of steps that were used in the standard reduction method to solve a 4x4. To begin with, we start by solving the white and yellow centre blocks (in generality, this can be any set of opposite colours), followed by pairing any 3 of the 4 white edge pieces, and placing them appropriately as part of the white cross.
After this, we move on to solve the remaining 4 centres, keeping in mind not to disturb the solves 3 white cross edges. Once this is done, the last white cross edge is solved, and then Edge Pairing is performed, which can be carried out in many different variations and mainly depends on the solver’s preference. Followed by this would be the 3x3 stage, here, since the cross has been solved already, one can directly proceed to F2L followed by the last layer.
One might wonder, why does this distorted sequence of the standard reduction method produce such an improvement in the speed and efficiency of the solves. Even if we think of an analogous example, isn’t it always faster to run along a straight path rather than run the same distance but along a zig-zag path? Even though as logical and correct as the above example solves, in the case of 4x4 speed solving (Where reduction is not considered a straight path), Yau is much faster, and let us see why that is so.
The first reason is the crazy level of speed and efficiency you get during the Edge Pairing phase. This is by far the single most useful advantage to switch to the Yau method. Prior to edge pairing, the white cross edges are already solved, this means that while edge pairing, we do not have to look at the bottom layer at all for any of the edge pieces, rather only at the top and the middle blocks. This single aspect significantly improves the edge pairing and saves more than 6x-8x of the edge pairing solving time compared to the standard reduction method.
Another important reason is that, once the edge pairing is done, there is no pause for the transition from edge pairing to the 3x3 stage, we can directly go to F2L without looking or figuring out the best way to solve the cross. This is also provides a major boost in terms of speed and efficiency, and if we can look ahead and figure out our first F2L during the final phase of the edge pairing, we can speed up the solve even further.
So, if we consider the previous analogy, it can be deducted that the Yau method more accurately resembles running along the straight path whereas the standard reduction method resembles running along a zig-zag path. The speed and flow are more pronounced in the case of the Yau method, and that is the reason why it is extensively used by most of the top speedcubers across the globe.