To illustrate Colin's point here are typical lines from a WIG file:

chrI 10 20 1.5
chrI 20 30 1.7
chrI 30 40 2.0
chrI 40 50 1.8

And these are typical lines from a BAM file:

ERR458493.552967    16  chrI    140 255 12M61232N37M2S  *   0   CCACTCGTTCACCAGGGCCGGCGGGCTGATCACTTTATCGTGCATCTTGGC BB?HHJJIGHHJIGIIJJIJGIJIJJIIIGHBJJJJJJHHHHFFDDDA1+B NH:i:1  HI:i:1  AS:i:41 nM:i:2
ERR458498.814362    0   chrI    1593    255 11M71474N40M    *   0   TTTCTACAAAGATTCTCTCACTTGTACAGAGGTGTCTTCCCATTCGTTTTC CCCFFFFFHHHHHJJGIJJJIJJIHJJJIJJJHHIJJJJJJJIJJJGIJIG NH:i:1  HI:i:1  AS:i:43 nM:i:2
ERR458496.427513    16  chrI    3782    255 51M *   0   0   CAGTAAAGGCTTGGTAGTAACCATAATATTACCCAGGTACGAAACGCTAAG JJJJIJJJJIJJJJJJJJJIJJJJJJJJGHBJJJJJJJHHHHHFFFFFCCB NH:i:1  HI:i:1  AS:i:50 nM:i:0
ERR458493.243111    0   chrI    3873    255 51M *   0   0   TGAAAATATTCTGAGGTAAAAGCCATTAAGGTCCAGATAACCAAGGGACAA @@<DDDDEH<<<CEFHAEIHHGGCHIGEEGG?@FF@<?ECGGGIIHIIIII NH:i:1  HI:i:1  AS:i:50 nM:i:0
ERR458494.816646    16  chrI    3972    255 51M *   0   0   TAATGAGCTAGTGATCCGGAAAGCTACTTTATGATGTTTCAAGGCCTGAAG @BBEBG;IGFDD@C:BHC<4A?94C<CH<EEEFABBFE<CDCBBD;DD=?1 NH:i:1  HI:i:1  AS:i:50 nM:i:0
ERR458497.736781    16  chrI    3976    255 51M *   0   0   GAGCTAGTGATCCGGAAAGCTACTTTATGATGTTTCAAGGCCTGAAGTTTG IJJJJJJJJJJJJJJJJJJJIIJJJJJJJJJJJJJJJJHHHHHFFFFFCCC NH:i:1  HI:i:1  AS:i:50 nM:i:0

As you can see, the WIG format simply contains some sort of value per (arbitrarily sized) windows of a given genome. These values can be, for example, normalized read coverages. BAM files, however, contain detailed information about every single read including a read's sequence, its alignment score and so on.