Standardized tests have wide standard deviations, and neither of these two practice tests is the real thing - so it's impossible to say which one is the better predictor.
The standard deviation on our test is 50 points; on the real test, it's 30 points. I don't know what it is for TPR (as far as I'm aware, I think we're the only company that has made our SD public). So the scores may not even reflect much of a difference - statistically, they may be quite similar.
It may even be the case that *both* are appropriate predictors even if they do represent a discrepancy. If, for example, you have a timing problem, then it's common for scores to fluctuate. Likewise if you have an attention problem that leads to lots of careless errors.
So, what do you do from here? You said that you'd just taken the MGMAT CAT. When did you take the TPR CAT? The more recent test is the better indicator (unless they were both taken within the past week).
Did you take both tests under 100% official conditions, including essay, IR, length of breaks, etc? If not, then the one that you took under more official conditions is the better indicator.
Next, how to get better?
Use the below to analyze your MGMAT CAT (this should take you a minimum of 1 hour):
http://www.manhattangmat.com/blog/index ... ice-tests/Also, read these two articles:
https://www.manhattangmat.com/blog/inde ... lly-tests/http://www.manhattangmat.com/blog/index ... -the-gmat/Figure out what you think you should do based on that analysis. Then come back here and tell us; we'll tell you whether we agree and advise you further.
(Note: do share an analysis with us, not just the raw data. Part of getting better is developing your ability to analyze your results - figure out what they mean and what you think you should do about them!)
Oh, and I mentioned that timing or mental fatigue might be factors - those are common causes of score fluctuations (if your numbers do represent a fluctuation). Look for timing issues during your analysis and also pay attention to the number of careless mistakes.