Wavelet based band pass

I read that wavelets are better for signal analysis in time-frequency. I saw that cwt() and icwt() can be used as band pass filters. For instance,

normal = %is a 1-D sequence data
fs = 300;
[wt,f] = cwt(normal, fs); 
y1 = icwt(wt,f,[50 100],'SignalMean',mean(normal)); % bandpass between 50 to 100 Hz

However, this results is slighty diffent than the results from bandpass(),

x2 = bandpass(normal,[50 100], fs);% bandpass between 50 to 100 Hz

In my research, I have to work with EEG bandpassed signals to build a high precision detector of abnormalities. Since I am new in signal processing I have doubts about these functions. Can someone explain/suggest me?