Study your past exam(s) and the solutions for that exam very carefully.
Are you missing problems that are similar to ones you have seen in workshop and/or the problem sets?
If you missed problems that are similar to previously assigned workshop and homework problems, you need to understand why. You should not be leaving workshop without understanding the problems there. Did you not understand the problems? Are you afraid to ask questions in workshop? If so, you should try hard to get over that. At least make a note of the problems that bother you and ask someone later. When you did the problem sets relevant for the test, did you try each of the problems on your own before talking it over with someone else? Are you relying on someone else to pull you through the problem sets in such a way that you really aren't understanding them? Make sure you give it a try first, then consult with others or the soln manual, but make sure you understand how the problems are done and why they were done the way they were. After the fact, you should sit down with the solutions to each problem set and strive to understand the hows and whys of each solution. You should ask someone about anything that doesn't make sense to you.
Are you making silly errors on things that you know how to do?
Try to get in the habit of checking over your work carefully. Do dimensional analysis and limiting case sanity checks. Does the problem seem excessively hard? If so, you may be missing a much simpler way of doing the problem. Take some of Manly's old exams for practice. Focus on not making those silly errors.
Are you having trouble getting the correct picture of the problem in your head?
Usually, Prof. Manly (or his helper) will answer questions pertaining to the interpretation of the problem. Ask if you aren't sure or if the problem seems impossible or excessively difficult.
Are you doing okay on things that are similar to what you have seen, but having real trouble getting started if you are asked something that is a bit different or new?
Often this might be due to relying on others too much while doing the early stages of problems in the problem sets. It can be due to lack of practice. Make sure you do all the problem sets and workshops! Practice the problems on old exams as if they were your exam. Often a problem such as this is due to lack of depth of understanding and/or lack of confidence. Both of these can improve with practice and experience. If you need more than the assigned problems, hit the books on reserve in the Physics and Astronomy library.
Are you extraordinarily nervous in the exam?
If you are having real issues with exam nerves on physics exams (more than you see in other courses), contact Prof. Manly about it.
Did you get a decent night of sleep before the exam?
Stay up with the course enough that you can get decent sleep before these exams. This isn't a memorization course. You need to be quick witted in order to think your way through new problems. DON'T CRAM this stuff!
How do you approach studying for the exam? Do you cram the material in the day or two before the exam?
Are you doing the problem sets? If not, that's the problem. You must do problems, problems and more problems to prepare well for physics exams! Physics does not cram. Try to stay on top of the course. It is a huge mistake to go thru the motions of doing the problem sets quickly without getting a deep understanding with the intention of getting that understanding just before the exam. Take the time as you go and refresh and review and practice problems just before the exam. Reviewing the solution sets is useful if done the right way (understand, don't just follow along).
How do you approach the problem sets?
Don't rely too much on friends or the solution set. Consult with others and/or the solutions manual after you have really spent some time trying to figure out the problem. Before starting the problems, review the material and read the chapter and look over the examples. When you first confront each problem, try to do it first without flipping back through the book immediately. Give yourself time to really think about it. If it does not work, write down your concept of the problem and your questions about it. Then go to the book, consult others, etc. Once you see how to do the problem, go back and try to see why you didn't get there in the first place.