It's less accurate, but makes it easier if two sides are disagreeing by a small margin since you can just sweep both numbers into -1 -2 etc
I see this exact argument made with the exact same wording every time the MU values are brought up.
the last 2-3 pages is why the ratio system is ******** in smash.
ITS CALLED A RATIO FOR A GD REASON.
65-35 is a ratio. that means that if you perform the same experiment 100 times, 65 times you will get the one result and 35 times you will get the other result. THATS WHAT A RATIO IS
How people can translate that to 'the person with the advantage should win 100%' is beyond me.
And since with all the variables in smash is not possible to narrow matchups down to 'how will this matchup play out if we test it 10/100 times' the way they can in games like Street Fighter, which is where the ratio system came from, it doesn't make sense to use it.
I think you're interpreting our MU ratios wrong. Smash ratios aren't
supposed to be read like SF ratios simply because Smash and SF are played very differently with very different goals in mind.
In SF, when Akuma:Blanka is 7:3, it means, "Akuma has a 70% chance of bringing Blanka's life down to 0 before Blanka does the same to Akuma." Therefore, you have Akuma winning 7/10 games and Blanka winning the rest. In Smash, X:Y being 7:3 should be interpreted, "X has a 70% chance of having the advantage at any one point during play. Y has the other 30%." Y may just take the lead or a stock during his 30% advantaged time or even during his 70% disadvantaged time. However, the chance of doing so repeatedly is notably less than 30%.
Now you see that the actual 'experiment' is being changed. In SF, the 'experiment' is who brings the opponent's health down to 0 first. In Smash, the 'experiment' is who has the advantage at any one given point in time during a match. Therefore, a single match actually collects a sample made of
dozens of trials. If a DDD beats a Pikachu during a certain match, but it was observed that, out of 100 situations during the match, he was at a positional disadvantage 35% of the time, then that MU can be assumed to be 65:35 in Pika's favor. This is why chars like DK and Bowser can win in MUs that have been delegated to be "unwinnable". Even though the ratio reads "95:5" or "100:0", that simply means that they are at a disadvantage all the time. That doesn't mean that they can't take stocks or the lead while they're disadvantaged, and it doesn't mean that they're doomed to lose 100% of the time. Likewise, a char with a 70:30 MU can reasonably expect to win more than 70% of the time. This is the problem I have with labelled MU values, such as +/-4 being "unlose/winnable". There is no such thing as an unwinnable MU.
Edit: I'll leave it to you to figure out what I mean by this, 'cause I'm too sleepy. In statistics, probabilities that are 'stacked' on top of each other are multiplied. Therefore, a 70:30 MU should equate very close to a 100% winning ratio in practical applications. If someone replies to this post, I'll expand on this. (-_-)
