Explain to me... how MLS and Complex Logic a/b aren't guessing?

I am not sure what you are asking but I'll try to give an answer.

I don't believe there is a consensus among the community regarding the connotations of words we often "hear" on these forums or in comments. To me, when there is mention of the word "guessing," it implies that a puzzler had to use absolutely no logic in placing a clue or part of a clue. The puzzler looked at the clue number, placed it in the row/column/etc., and leaves it to lady luck to hope there is no future conflicting clue.

When puzzles reach a point where there is no apparent next move, advanced technique/complex logic like MLS must be used. It is more of an assumptive approach meaning you are essentially making a temporary guess to deduce whether or not an error or advancement can be made. If no error occurs, then it means a possiblity and you must try elsewhere. If you can ascertain that the assumption does not work, then usually it ends up allowing one to place background or some other color. If one finds that using advanced technique results in no significant advancement, then one probably isn't using the technique correctly. Almost all advanced techniques require some mode of assumption, but it always requires logical deduction.

Sometimes it is arduous and time consuming but I for one would rather be certain that I can advance correctly than leave potential errors riddled in the puzzle as I progress. It takes an extremely talented puzzler to simply guess and if a mistake arises, actually be able to know how to fix it. This skill takes a lot of experience.

I hope I answered you or at least made some sense.

There are often patterns in the numbers. They're most obvious on longer thinner objects, rather than (non-simplistic) blobby objects. With multiline solving, you can look at the patterns of numbers and see where a 4 in one direction is composed of a 1, 2, 3 and 5 in the other direction, and can see where the partial lines above are completed by the numbers next to the 4. This isn't intuition or guessing, or else I'm psychic. The more reasonable explanation is that the patterns I think I see are real. Once in awhile I'll follow the wrong string of numbers and make a mistake, but I mostly do large multis, and do puzzle after puzzle with the patterns working out just as I expect them to. That is, many a point value in the millions is dead easy for me because I'm following the patterns, but hard for the algorithm because it can't see them.

That said, my least favorite kinds of puzzles are complex 3 color and blobby two color (i.e., B&W). Simpler 3 color can still be solved by inspection and number patterns. The complex ones mostly only come together (at least for me) with marking out the places where one of the colors logically could not be until the only open areas just fit the numbers. Some people love these for the challenge, but I find them annoying because I want a puzzle to relax with, not to have to think deep thoughts about. The blobbly two color ones also don't have any recognizable patterns, or not enough, so one has to count out each line and move back and forth like the compiling algorithm does. That's not hard, but I don't find it fun. You can do higher order MLS on them, where you do a meta-analysis of the puzzle and put stops in the logical holes, but again, not restful.

In the really really hard ones of both these types (i.e., especially hard rather than just annoying), the only way I've found to solve them is through elimination of possibilities. That might be what was looking like guessing to you. It's not so much guessing, as assembling the possible and impossible then trimming away all the ones that don't work out as you add more information to your matrix. That's still logic, not guessing, but, in a way, you do have to assemble all the positive and negative guesses in order to make your decision tree that leads you to the right answer. Really really not restful!!

Try an 8 color multi. You'll see better, maybe, how MLS works by logic rather than guessing.