TSON, my issue is not centralization or brokenness. It's a move basically either doing its normal effect... or killing you. Again, I ask: have you ever been hit by the PK flash effect that does...
nothing? It's just outright ****ty design.
My personal opinion: there are in fact some cases where randomness is justified. Things where the randomness is an inherent balancing factor or leads to "cycling" (think Peach's Turnips or DDD's waddles); and in some cases you simply can't effectively separate the randomness from the move (G&W's Judgement, the aforementioned two) without a complete redesign. That's not in anyone's best interest.
However,
purposefully injecting randomness into the game should be avoided. What if Wolf's laser occasionally did 50% with no warning whatsoever? What if Warlock Punch came out randomly in 10 frames instead of 90 every once in a while? What if... You get the point. Obviously, neither of those random effects are game-breakers. Just like Ness's PK Flash occasionally OHKOing vs. occasionally doing absolutely nothing is–it's not a move you're going to use a lot, and it's clearly not a huge deal that it's random. But it still inserts a random element into the game-one which is detrimental to competition... and poor design altogether! And then there are the ones that
actually matter. Imagine if in a tournament match you have a solid stock lead and then suddenly you get hit by the OHKO counter (the normal counter wouldn't have come close to killing you). Or if you land just out of range of ROB's fsmash, ready to counterattack, and he gets the laser effect-completely randomly, you just missed your punish, and maybe even died.
This is another issue where I think it's not diluting Minus's formula at all, but it's just
****ty game design. For reference:
this is what I'm talking about. Why did we add randomness? Canon? I have never even
seen a mother game; for all intents and purposes, the Melee and Brawl versions of PK Flash
are canon. ****s and Giggles? Yes, because a random 1/25 chance of a OHKO is
a really ****ing good idea. I don't want to make this game competitive to the exclusion of casuals, but
this is the wrong way to do it. It's like with tripping-sakurai wanted to make the game accessible to casuals, but guess what:
everyone hated it, including the casuals.
This is my stance, and I support removing quite a bit of the randomness currently present in Brawl-.
EDIT: I'm going to quote the team's latest addition, "TheEffinBear" (for he is awesome):
The best source of randomness is the players themselves, because people are so very bad at being truly random. The interplay between these patterns and actually predicting them, which leads to predicting your opponent's predictions and changing your patterns to counter your opponent's predictions, which leads to etc etc. There's a LOT of gameplay that comes out of prediction, and innate randomness in a move throws a lot of that out the window.
So yes, for the sake of competitiveness, you definitely should want to remove or at least minimize randomness.
But another aspect of randomness is very very undesirable: the frustration it causes in its victims. Valve has put a lot of work and analysis into TF2 for making it enjoyable experience with as little frustration as possible, some of which they explain in this blog post: <snip>
What I really want to focus on is:
* "...for death to be a positive experience, players had to feel like they could have avoided dying if they'd done something different. [...] If you don't know what killed you, that death is failing in providing you the feedback it's supposed to, and you won't be able to figure out what you could have done differently. Unsurprisingly, we saw that these deaths were highly aggravating to players, and in sufficient number caused new players to stop playing entirely. Trying to reduce the number of these deaths in TF2 was done through a variety of changes. It was one of the reasons why we chose to remove the hand-held grenades that each class had in TFC, which were one of the primary causes of these deaths."
This directly applies to a lot of what Minus does with the increased mobility, which helps you engage your opponents (likewise for my Samus 2.0 changes which help remove her 1.6 keepaway game that just obliterated certain characters). But it also applies to randomness, especially the part about TFC's grenades: if you and your opponent reiterate an identical sequence of moves with identical spacing, etc, but one of those iterations results in you being OHKO'd, that's quite frustrating. Or from the other end, if one of those iterations results in your move just failing to do anything, that's also quite frustrating.
For an example of HOW frustrating this kind of randomness is, harken back to the earlier days of vBrawl. Compare the incredible volume of complaints about Ike's "combo" of "opponent trips -> f-smash" with how many times that ACTUALLY happened. How often did a random trip actually change the outcome of a match? It's not necessarily rational to hate the random tripping as much as all that, but the fact remains that the community LOATHED and practically revolted against it (wasn't "No random tripping" like the first hacked Brawl code ever?). Engaging your opponent is one of the very foundational things that a fighting game IS, but a random trip rips you out of that situation and forces you to suffer the random number generator's whims and punishments for a moment -- these trips are "highly aggravating to players, and in sufficient number caused new players to stop playing entirely".
Now, I grant you: moves with innately random components are significantly less frustrating than the random tripping. But the same elements are present for that same type of frustration: eating a random-chance OHKO in a fighting game doesn't typically make players "feel like they could have avoided dying if they'd done something different". It feels like a fluke (and as a random-chance OHKO, it is!), not a moment of having been outplayed and an opportunity for improvement. It tends to drive players away rather even if the average risk and average reward are perfectly balanced.