I'm not a physicist, but as far as I understand, it just doesn't work like that.
Entangled particles start off in a superposition of possible states, when we measure the state of one particle, we can deduce the state of the other particle because we know that the superposition collapses for both particles when the state of either particle is measured, however, no classical information is actually communicated between the particles at the time of measurement.
Just like how the the collapsing wave function in the double-slit experiment appears to "reach back in time" to destroy the interference pattern, a similar effect occurs when the superposition of our entangled particles collapse. Scientists have ruled out the idea that a "hidden variable" secretly stores the true state of the entangled particles prior to measurement, but for the purposes of communicating classical information, this might as well be the case.
Think about it this way.
You create two books with a table - in one column, the time, in another column, a value, either 0 or 1. They are identical, except that the 0s and 1s are switched in both books.
Now you give one of the books to a space traveller, who gets on a rocket that will take him 10 light years away from Earth. You also agree, that once he is there, both you and him will look at your respective books and look at the same time and it's value. So he goes away, and then when the time comes, you open your book, look at the time and read it's value - it's 1, so that means that his book must say 0. Now, there is no "transfer" of information involved - you have not communicated with someone 10 light years away. The information was encoded in the book when it was created, not when you read the information. The same with quantum entanglement - it's already decided when the particle is created,not when you look at it - so it doesn't matter how far the particles are apart, you are just reading pre-defined information.
Edit: no, I am not a physicist either, but I read this explanation somewhere and I think it explains it fairly well.
Entangled particles start off in a superposition of possible states, when we measure the state of one particle, we can deduce the state of the other particle because we know that the superposition collapses for both particles when the state of either particle is measured, however, no classical information is actually communicated between the particles at the time of measurement.
Just like how the the collapsing wave function in the double-slit experiment appears to "reach back in time" to destroy the interference pattern, a similar effect occurs when the superposition of our entangled particles collapse. Scientists have ruled out the idea that a "hidden variable" secretly stores the true state of the entangled particles prior to measurement, but for the purposes of communicating classical information, this might as well be the case.