Meaning of probability in a multiverse/a many-world interpretation?

Consider me tossing a coin and I got tail as a result on observing it. Then, what would be the result of the 'parallel me' in another universe? If the 'parallel me' gets head as a result then, considering all the parallel universes the probability of occurrence of any event would be 1! In that case the whole concept of probability (considering all the parallel universes) goes wrong. What's your opinion about this or am I thinking in the wrong way?

(I actually asked this question while participating in 'I'm a Scientist get Me Out of Here' but I would also like the views of Physics SE members.)

• There's always a few universes where the coin lands on its edge:-) – Jitter Nov 17 '13 at 17:10

The sum of the probabilities of all possibilities being equal to 1 is necessary for probability to be well defined. In other words, I wouldn't have it any other way.

You should specify what type of multiverse you're talking about.

Anyway, I don't see the problem. The important thing is that the probability is lower than one in each universe. So probability is useful en each universe.

Don't let any tell you there would be no problem.

In a MV snake eyes would come up the same as 7's rendering probability meaningless. In fact a few physicists have submitted papers that propose the universe must be destroyed soon so we can calculate finite equations. Thus fixing the odds. They seriously predict it must end soon or nothing will make sense.

This is the price of rejecting the fine tuning as actual to avoid its obvious conclusion that there is a designer.

• Hi, welcome to Physics SE. I think you have a valid point but this post might not be comprehensible to someone who has never read about the topic. Could you maybe elaborate on your point and include the keywords and links relevant for this discussion? – Void Sep 21 '14 at 9:34
• "In fact a few physicists have submitted papers that propose the universe must be destroyed " Or, you know, MW is wrong. – Bubble Nov 17 '14 at 20:28

In a multiverse, everything happens (at least in the many worlds view you seem to be talking about). Probability is applied to bundles of trajectories through the state space. Say you are an observaer at point P in this abstract parameter space, then if you specity a hypervolume in this space you could, in principle, get the probability of being in one of those states (i.e., seeing heads, tails, getting to work on time, NOT getting to work on time) etc. Of course, we don't have the necessary wavefunciton to actually do this, I have no idea what thse calcs would look like, but that is how I think of probabiliy in QM.

Why don't you try to do the experiment, and then ask your other "me"s what results they get. If they won't give you their results, I would simply forget about it, and just accept my own observations to be the only reality, and the rest just fantasy.

Following the the toss of a coin with probability of heads of $p_H$, $p_H$ of the universes branching from the point of the coin toss will have "heads" occurring, while $1-p_H$ will have "tails" occurring — that is, following the toss there are two families of universes with the toss in their past and occurring in proportions $p_H$ and $1-p_H$.

What you see is determined by which of these families of universes you "end up" in, and that is random: you will end up in a "heads" universe with probability $p_H$, and a tails universe with probability $1-p_H$. The same is true of all other "yous": $p_H$ of them will see "heads" and $1-p_H$ will see tails.