For what it's worth, I HQ 2star all day erry'day but never use Rumination (since getting BB at least, admittedly not everyone has that advantage) and never use Manipulation or Mend on anything but 40 dura synths (where it's basically unavoidable to get your IQ stacked). At this point I don't even TotT on 1star items anymore - though thats more to do with my gear and ease of HQing 1star mats at this point more than anything else.
The value of WN is lost in the obsession with the "cp / durability" measure, which is useful but one-dimensional. WN is valuable because it increases the potential value of each durability two-fold, so even if it results in a net loss in durability compared to Manipulation, that durability has been twice as productive. They're two paths to the same goal, but to me all that matters at the end of the craft is how many times I can/need to hit the damn thing, regardless of how much raw durability it takes. With that said, I think working off of "CP per 'action'" is much more useful. Or even, "CP per 'additional action'" which would be [CP Spent (on WN/MM/MM2/MAN)] / ([total strikes taken] - [max durability / 10]).
This is best illustrated by using an 80 durability craft, as you simply do not have enough room on 40 dura for WN to show it's longterm value beyond the start of a craft; though I always always always start 40 dura crafts (well, high level ones) with IQ/SH2/WN/4x HT.
For example:
80 dura craft process using WN v MM2 (widely cited as the most efficient use of CP to durability)
IQ - 18
SH2 - 25
WN - 56
"Action" x4
SH2 - 25
WN - 56
"Action" x4
SH2 - 25
WN - 56
"Action" x4
This will drop a 80 dura craft to 20/80, give you 12 "actions" and cost you 243 CP to do so. 243 / 12 = 20.25 CP per action.
IQ - 18
SH2 - 25
"Action" x5
SH2 - 25
"Action" x1
MM2 - 160
"Action" x3
SH2 - 25
"Action" x3
Using Masters Mend 2 this process will drop an 80 dura craft to 20/80, give you 12 "actions" and cost you 253 CP to do so. 253 / 12 = 21.08 CP per action.
However it has the benefit of having 2 SH charges 'left over' whereas WN would require you re-up SH. How much this matters is debatable considering two SH charges is likely not enough for you to completely finish both your quality and durability on any craft you would actually use this approach on, thus probably requiring you to re-up SH anyway.
TotT arguably throws a wrench into things and favors the latter technique because you, of course, aren't losing out on the WN charges; but if you just look at the net result there's not a huge difference. The CP differential between the two does not change (it only shifts) and the only tangible difference is actions that will then occur in/out of WN and how many total actions are possible. This is assuming the "Good" occurs during a WN of course - goods between WN should absolutely be Tricked.
TotT taken during WN // Delta durability after WN // Delta actions after WN // Delta actions in WN // Net action change
0 // 0 // 0 // 0 // 0
1 // +5 // +1 // -1 // 0
2 // +10 // +1 // -2 // -1
3 // +15 // +2 // -3 // -1
4 // +20 // +2 // -4 // -2
5 // +25 // +3 // -5 // -2
etc.
So you might lose up to two actions in normal circumstances, as such your "cp per action" will increase and be inferior to that of MM2. However I think there's one more unsung benefit to WN, and that is granularity and how it helps mitigate randomness.
With MM there is no middle ground, you can't take "2/3" of a MM2 even if that's all you would want/need (though you could use Manip/MM I suppose) - but you can do only 2/3 WNs (which is actually what I would advocate - three is overkill unless you have absolutely atrocious luck with HT). And what this does is open up even more CP ( 160 MM2 - 112 2x WN = 48) which is important because of how it enhances your options for your 'finisher'.
I always want to make sure I have enough for my finishing combo and spending ~112 CP on two WNs instead of 160 on MM2 makes that easier, especially if you aren't using CP food or don't have great gear/materia. You can accomplish the same thing (100% quality) by simply stacking your IQ higher with the additional touches and using a 'lesser finisher', true (again, two paths to the same goal) but I want to need as few successful touches as I possibly can to mitigate the effect bad luck can have. So to me, knowing that I 'only' need X IQ stacks + my 'full' finisher is superior to X + Y IQ stacks with a 'cheaper' finisher because it increases the amount of quality that is not left up to chance.
Ultimately I think on 80 dura, you're splitting hairs between MM2 and WN. You've got enough room either way to be OK with the proper skills imo. But I absolutely prefer 2x WNs to 1x MM2 because of the CP savings that I can then allocate towards my SH/Ing/GS/Inno/BB finale. Admittedly if you've got the CP materia and food, you wouldn't really have a problem reaching that point either way. But for most people who are much closer to 300 CP than they are to 400, I think 48 can make a big difference to how you finish your craft, especially since the MM2 process as is would take 253 out of most peoples (probably) roughly 300-330 CP.
==================
On 40 dura, WN is still very useful but less so because it needs to be alternated with a Durability restore of some sort to prevent a break.
40 dura with WN and Manip
IQ - 18
SH2 - 25
WN - 56
"Action" x4
Manip - 88
SH2 - 25
"Action x3"
20 dura with 212 cp spent on 7 actions: 212/7 = 30.29
I use the above method for all my 1star mats and HQ almost every single one (and for my 2star mats, but with them I actually use TotT and CP food and go for 2x WN and a Manip with Comfort Zones used as well (the target for me is to always have 147 CP at the end of the craft)). The advantage here being (again) a stronger finisher compared to one more swing (and a trivial difference in cp/action). I finish every one with SH + Ingenuity 2 + Great Strides -> If Good (Byregots Blessing) else (Innovation) -> Byregot's Blessing -> Careful Synthesis 2 (and end up with 0 cp or 2 cp with no food? I forget exactly).
Most crafters could accomplish something similar with SH -> Ingenuity -> Great Strides -> Advanced Touch -> Standard Synthesis at a cost of 135 CP, meaning a starting CP of 337 - whereas doing the same with Manipulation only (below) would require 372. But, I'll get into this below, unless you're using Byregot's Blessing you should almost always end with your Rumination returned CP - probably making this less of an issue.
40 dura with Manip only
IQ - 18
SH2 - 25
"Action" x2
Manip - 88
"Action" x2
SH2 - 25
"Action" x1
Manip - 88
"Action" x3
20 dura w/ 244 cp spent on 8 actions: 244/8 = 30.5
So essentially same CP cost per action but you spend more CP at once leaving you with less available at the end. Again my preference is to spend less CP on the process and more on the finisher but you really need to experiment based on the skills and gear available to you - and be able to adjust on the fly if you have particularly bad (or good even) luck.
Ultimately I think the difference maker is Byregot's Blessing. Without it, you pretty much always have Rumination available to you at the end of a craft for a pretty significant CP return, so saving throughout the craft is not as vital - mitigating some of what makes Waste Not valuable. But when you have BB, I think you can use it with WN to make your HQs more reliable and faster by spending less CP on more Hasty Touches and allocating it towards more 'guaranteed' quality through your other skills (or even Basic Touches).