Sealed versus vented

The choice is usually presented as a matter of taste. It is mostly a matter of which compromise you want.

Sealed

A sealed box is a second-order high-pass: 12 dB per octave below cutoff. The trapped air acts purely as an added spring, and the whole response is set by one number, Qtc.

• Gentle rolloff, which often integrates well with room gain
• Least group delay, best transient settling
• Smaller box for a given driver
• Cone is loaded at all frequencies, so no subsonic hazard
• Less efficient, and less maximum output near cutoff

Vented

A vented box is a fourth-order high-pass: 24 dB per octave below tuning. Near fb the port radiates and the cone nearly stops.

• Around 3 dB more efficient for the same size and cutoff, or a lower cutoff for the same size, which is Hofmann's Iron Law stated as a choice
• Much lower distortion near tuning, because the cone is barely moving
• Steeper rolloff, roughly twice the group delay near cutoff
• Larger box, and a port that has to physically fit
• Cone unloads below tuning: a real hazard needing a high-pass filter; see subsonic unloading

Which the driver wants

The driver has usually decided already. A low Qts driver suits a vented box and will sound thin sealed; a high-Qts driver suits a sealed box and will demand an unreasonable enclosure vented. Around 0.4 either works.

An honest note on "sealed sounds tighter"

The claim is common and is partly real and partly not. The real part is group delay and the behavior below cutoff: a sealed box rolls off gently and stays controlled, while a vented box below tuning is unloaded and can be driven into audible trouble by material the designer did not anticipate. The unreal part is the suggestion that a correctly aligned vented box is intrinsically slow or boomy in its passband. It is not: a poorly aligned one is, and there are many poorly aligned ones.