Stop being silly. If your proposed Item had a 100% chance of killing an enemy Demigod, would that then make it better to you??? Of course not. I`m not even going to try to engage that. If you have a point to make, try utilizing a better example.

As for L4D, some people contended that randomization has no place in tight competitive games. I cited countless examples of randomization core to Demigod, as well as L4D`s seeding of game elements. Now that I`ve made some logical points, suddenly you are parsing the argument by trying to delineate between games wherein you compete and those where you *really* compete... .

(frown)

I`ll try again.

In football you have referees. If you are lucky they make good calls. If you are unlucky they make bad ones. Does this trait of chance then make the sport of football a poor competitive game?

With Demigod we are talking about a few underlying mechanisms.

Anyway, believe what you wish.

Let's look at some of your examples:

It's true, there are random elements in competitive RL sports.  There are random elements in competitive virtual sports.  But in none of these competitive sports do the mechanics of your *own* abilities depend on chance.  That's a crucial distinction, and I don't think it's splitting hairs to say that your own abilities, no matter how broken, cannot be dependent on chance if you truly want to measure skill between players.  

You see though, the problem is that stuns as they are with a 100% activation rate is just as broken as an instant death item.  It isn't a streatch.  Here are the facts.  Getting stunned leads to a death that cannot be avoided by any user actions.  The Instant Death item leads to a death that cannot be avoided by any user actions.  Both stuns and Instant Death are broken with a 100% chance to activate (hence the existance of this thread).  Reducing either element to a 50% chance would leave both elements broken and turn the entire game into a crapshoot.  Thus, they are actually quite similar in effect.

I never argued that L4D's random elements were wrong.  For what the game is, it works extremely well.  However, L4D and Demigod are two entirely different games.  You are citing a card game to support the inclusion of randomization in chess.  It doesn't work.  One thrives with randomness and one dies with randomness.  Imagine trying to play chess, but being forced to watch your Queen have a heart attack and die on the second turn.  You might as well give up.

Can randomness work?  Yes, in the proper game.  Is Demigod the proper game?  No.

If you take your favorite professional football team (name not stated to avoid a flame/off-topic war) and pit them against your average college football team, who would win?  Even if the refs make every single call in favor of the college kids, do you think it is at all possible for the professionals to lose?  No, the professionals will win.  Why?  Because the randomness is not significant enough to bridge a massive skill gap.

If you take an amazing, professional stunlock team in Demigod and pit them against an average stunlock team, the amazing team should win.  However, if you reduce stun chance to 50%, then the amazing team can rely on nothing.  The key difference is that stunlocking is the strategy.  The above stated game would turn into a crapshoot.  Sure, it would be much more difficult for the scrubs to win even if they hit every stun and the enemy misses every stun, but they could still do it.  This is very unlike random elements in football, where such a skill gap would never be overcome.  The element of randomization in this case makes two criminal actions.

1.  The degree of randomness is enough to skew any game by a highly significant amount.  If you miss three stuns and they hit three stuns, they may have just gotten you down the slippery slope.  GG, even if you may be better.

2.  The randomness is introduced into user defined actions.  A ref's calls or the passive item boosts in Demigod are things that sit on the sidelines.  They exist and influence the game, but it does not stem from a user action.  User actions need to be responsive.  If you ask the average player why they hate stuns, they will say because it takes control away from the player and renders him helpless while he is finished off.  Your solution may theoretically solve the cause (giving you the benafit of the doubt here), but in doing so it will reproduce the symptom that people hate.  You will remove responsiveness from the game by making user actions sometimes worthless, thus taking control away from the player.  Loss of control is the reason people hate stuns, and following that logic your solution is just as bad as the problem it is proposed to solve.

No, of course not.  Football is a competitive game DESPITE the fact that it requires human referees that are prone to mistakes.  The same can be said for any sport that requires officiating.  Basketball is a great example of a game with honestly terrible officiating (to be fair it's really tough to consistently make good calls in basketball) which is nonetheless very competitive and fun to watch.

All of the same things can be said about the other factors you mentioned - weather, potential for player injury, etc.  Those are necessary evils that real-life games have to deal with because they are in real-life.  We are playing a video game in a virtual sitting in which randomized mechanics which hurt the competitive nature of the game can be minimized to the greatest extent possible.  For your reference, competitive means in an ideal world the best player or team wins.  Random chance allows for the possibility of the worse player winning, which from a strictly competitive point of view is not a good thing.  Underdogs are great, and unexpected victories and miracles are great too.  But let's keep them on television and on the football field / basketball court / baseball diamond etc. and out of gaming.

So if I have the right combination of Items, my Demigod can play with a 50% Evasion/Dodge. On average every 2nd strike, including special ability strikes, misses. To some of you this is a trivial (?) trait.

So bereft of a 100% stun capability, good Demigod players will have nothing to do, nothing to use (near quote).

Hmm.

All players work under the same rules. Randomizations apply to all.

[The Close Combat series - another example of a competitive experience suffused with randomization mechanics.]

Anyway, I have other things to write. Believe what you wish.