Your code doesn't show how data are being acquired; is it in lots of 100 hundred values, vectors, arrays, ...?
You would have much better success to use arrays and indexing rather than all the separately named variables, then calculations could be done in a vectorized form using the abilities of MATLAB. Passing those arrays to functions inside the looping construct that returns the data would also greatly simplify the code structure.
In "I was using (end+1) to add aditional values onto an array but this does not go back and recalculate std values for prevous shots" it isn't clear why you would need to "recalculate previous shots"; those data aren't going to have changed so those previously computed std values will be the same and already saved -- unless you mean you want to be calculating a moving std or other measure instead?
Also in the above, using (end+1) to add elements is very inefficient as it requires reallocation of the array every pass. This may not be terribly bad in extra computing time for small arrays or if the number of loops is few, but is bad practice in general. Instead, preallocate the result arrays to some expected size first and then increment an index into them and store the values there. Test for the process running enough times to fill up the array; if/when does, then you can reallocate another chunk onto the existing array and continue; this is much more efficient.
But, the big thing is to factorize the calculations into a generic subroutine and pass the incoming data into it as arrays, thus eliminating the need for all the many separately-named variables.
Providing an overall description of the actual set up and the expected result would aid in folks here being able to help you refactor the code rather than trying to guess from incomplete code snippet.
