It's worth noting that we're still arguing two different things.
First of all, let's address the notion of bottoming out being wherever you define it to be. Your argument is there's no difference between hitting the bottom that doesn't move and the bottom where your force would normally bottom out. Except there is. There's more force needed to hit bottom than to hit halfway. That's typically how springs work. The more you compress a spring, the more force required to compress the spring. If you're bottoming out vs not bottoming out, that's not the same amount of force being applied. This is pretty obvious for linear switches but gets more complicated with tactical switches based on the severity of the tactile bump. Unavoidable on a buckling spring. Typically for an mx style switch though, after the bump there's additional force that would need to be applied to hit the bottom. Admittedly, on the HP graph, the top of the bump is roughly the same as the bottom of the switch which would lend itself to your notion that it's the same thing. Not the case across the board. There's no sense arguing this point. We can google the force graphs for each switch and see the data instead of arguing opinions. Want a clear example of my argument? Look at the Cherry MX clear force graph. Far steeper than that of a brown.
I will say that when you add o-rings I will agree you are effectively changing the bottom out location. But they are still acting as springs though. If you press hard enough, the switch can still hit bottom once the o-ring deforms enough. Personally, crushing the key to the bottom or wherever it will let me go doesn't make me any faster. If that's not the case for you, it's not the case for you. Who am I to argue? I don't know how you type. If you're henpecking vs touch typing, I don't have a leg to stand on. If you're touch typing, maybe I do.
And I also explained in my previous post that you have to type without bottoming out wasn't my intent. But the distance traveled to bottom matters. If you type equally fast on a switch with 4mm to bottom out and 2mm to bottom out, that's additional 2mm each finger needs to travel. My suggestion is it's not necessary to press the key all the way to the bottom to register a keystroke. If we have equally fast race cars and I have to travel half the distance to the finish line, I win.
Analogy: If we have equally fast racecars and I have to travel half the distance you do, I win. Also speed and power are often tradeoffs. A heavyweight boxer is not as fast as a welterweight but has much more powerful strikes, do you agree?
Maybe my analogy doesn't hold up when it comes to typing. Hell, if you search on GH there's a thread back in 2011 where they're discussing bottoming out vs not bottoming out where there are people agreeing with me but also people saying if it happens it happens, doesn't matter. Same on Deskthority. Same on Reddit. Sounds like the jury is out so at best and we're arguing opinions and personal preference.
Math: F = -kx where k is the spring constant, x is the travel distance, and F = force. Assuming a linear spring constant (haven't a clue what the spring profile is for mx style switches), if x = 2mm vs 4mm, twice the force would be required to depress it that distance so F (me) vs 2F (you), right? Since F=ma, assuming the finger, switch, or keycap don't change in mass, this would tell us with twice the force applied to the same mass, you'll have twice the acceleration. s=Vi(t)+1/2(a)t^2; your distance of 4mm is twice 2mm so for you it would be s(me) = 1/2 s(you). S=0+1/2(a)t^2 (for me), 2S=0+1/2(1/2a)t^2 (for you). Let's hold a constant and modify it by saying (1/2)a. So for you it would be 2s=1/2(1/2a)t^2 --> 2s = 1/4at^2 --> 8s/a = t^2; t=(8s/a)^.5 (for you) and s=1/2(a)t^2 (for me) so 2s=at^2 --> 2s/a=t^2; t=(2s/a)^.5. Can we agree that (8s/a)^.5 > (2s/a)^.5? So t(you)>t(me).
I'm many years out of school and this is extremely simplified math (assuming a horse is a sphere to make the math easier), but the principle is there. There's a lot of gaps in my math here in terms of factors such as resistance, the fact that once you depress the key, force is still required to retract your finger, aerodynamics, chemistry (C8H10N4O2 levels, eg), how well you slept, shoe size, how long it's been since you at Chipotle...you know the drill. Honestly, I was going to omit the math but after taking the time to remember, I'm leaving it, haha. I will be happy to revise if anyone wants to add or correct.
tl;dr - Bottoming out doesn't make me faster and I feel like it can be quantified even if I can't qualify it. If someone else finds bottoming out to be faster based on their typing style, then that's faster for them. At the end of the day, OP has two differing opinions and they can try them both out and see what works better for them. And maybe we can offer differing opinions without calling someone else's argument BS because it doesn't apply to them even though there's multiple threads on here, DT, and Reddit that have people on either side of the aisle.
I'm sure once I get around to videoing and uploading there will be plenty of criticisms of my argument because you made me question myself so I gave it a shot and hit 112 while only bottoming out space bar and backspace regularly with the occasional double letter getting a bottom as well. Perhaps we should both upload a video so we can critique each other's typing styles too?