how do i tell what the fibonacci ratio for a stock is telling me?

The concept behind the Fibonacci ratio is that many natural processes uses the Fibonacci sequence as a sort of natural pseudo-random number generator. Trees will use the Fibonacci sequence to determine how many leaves and branches to grow. You body uses the Fibonacci sequence to determine how capillaries spread out and form a net. The proportions of your body is determined by the Fibonacci sequence. Hence the Fibonacci ratio is characteristic of things that appear natural and organic to us. However, there is nothing natural or organic about the stock market, there is no basis to believe that the Fibonacci sequence is fundamental to stocks.

The Elliot wave is the assumption that the stock market is a fractal with the basing, growth, plateau, and decline phases of a product life cycle as it's fundamental unit. However it makes the assumption that each fractal is an integer multiple of the other and hence is a harmonic. It's like saying that white light isn't all the colors in the rainbow but only the ones that are harmonics of each other. Also, the proportion of the basing phase to growth phase, plateau phase and decline phase are also set in the Elliot wave and we know that isn't the case, some things take a long time to get off the ground, others take off right away, some things have a lot of staying power and decline slowly, something fall right away. Fundamentally, the Elliot wave is flawed because of the assumptions that it makes.

You need to use year's worth of data, ten years may not be enough, one of the criticisms of LTCM was that they only used four years of data.

Support levels are the belief that there are prices that a large number of people are likely to begin buying into the stock at. Good reasons for a support level are a rationale for value at that price i.e.: discounted cash flow, and historical evidence of support at that level. Bad reasons are the extrapolations such as Fibonacci and Elliot wave. The Fibonacci and Eliot wave suggest that there may be support at those levels if the underlying mechanisms were that of their respective fractals but you know with certainty that the underlying mechanisms do not correspond with Fibonacci or the Elliot wave. You can't trust either levels but you may be able to bound your chart with them to see if there is any historical evidence of support in those ranges. For example window the data to windows where the data enters the range and exits the range and do linear correlations over those windows then compare those to linear correlations of windows from where the data left the range to where they enter the range. Those would be support levels if the linear correlations of the windows where the data is between the bounds are significantly flatter than outside and much more correlated, the data outside the ranges should also have flat linear correlations but low correlations. They are not support levels if the respective windows of data within the bounds have various slopes and high correlations. As the bounds represent a significant smaller subset of the data, you would not expect the windows of data between the bounds to have low correlations compared to outside the bounds, if so then the bounds are too broad.