The article deals with the results of mathematics examinations at the University of Finance and Administration in Prague before, during, and immediately after the COVID-19 pandemic-related restrictions. The first objective is to evaluate whether the non-standard forms of testing (correspondence and online), used on an emergency basis during the pandemic, were adequate compared to the standard form (face-to-face) applied before the pandemic. The second objective is to assess whether and to what extent the results of the examinations have changed after the return of teaching and testing methods to normal. It turns out that the use of non-standard forms, although more challenging for teachers to control, did not lead to better results -- the results in the correspondence form were similar to the standard form and even worse in the online form. The results of examinations administered in the standard form after the return to normal teaching were significantly better than in any of the periods studied, including the standard form of examination before the pandemic. Possible reasons for the results are analysed in the paper.