My initial difficulty is that 'decent' is subjective. I have lots of personal stuff (going back decades) 'mixed' for personal review (via cassette, CD4Automobile, hand held flash drive player, etc.) On which nothing was adjusted from original, (if levels are set correctly why would you need to alter level or pan in post?) . . . I started tracking live performance with a mono cassette pretty much as soon as portable boxes were available. As soon as 'tape out' was added to console I used that and as I picked up consoles with additional buses graduated to 2 track recording based primarily on 'house' sound with a mix that compensated for elements not supported adequately in the PA (Bass typically DI'd, etc.). These 'mixes' had mechanical level adjusted in front of tape primarily because that was/is approach used to create the performance sound. I've used mono cassettes, stereo cassettes, stand alone DAW's, hand held flash drive recorders to record performances that used no electric or electronic sound reinforcement, but even here you're still controlling performance via the mechanics (piano damper and sustain pedals for example, mutes on horns) of 'transducers' referred to as 'interments'.
Generally speaking if the budget supports it I tend to track 'dry' and edit in post because that presents the greatest range of options. And a lot of the standard tools can be used subtly to balance harmonize elements of a performance in ways not dissimilar to moving to front of stage, moving to the side of the ensemble. There are times when compression is a subtler and perhaps 'better' approach to balancing level then touching a fader, there are times when time based effects are as reasonable approach to harmonizing voices (instruments, elements) as twisting or sliding an EQ control.
But what constitutes a 'decent' mix is dependent on goals of final distribution. A 'mix' I might create for review by me or band might not be appropriate mix for paid or even gratis distribution to 'audience'. Tracking as neutral, i.e. Dry 'un mixed', as possible permits the greatest number of options on how to deploy the performance. But everything you do, everything you use, certainly the room, impacts, non-linearly, amplitude, frequency, dynamics (ASDR), time of individual waves as well as the 'blend'.
I guess main point is why, except for very specific & limited goals (a la John Cage for example), one would want to impose limits not use dynamic and time based tools? If you simply allow dynamics to be chaotic, pseudo random, then balancing amplitude is going to be that much harder. Humans, biologically use the mix of direct and reverberant sound for 'location' so surrendering control of reverb makes 'panning' far more difficult (to impossible). We speak of 'bass' waves as being less directional then higher frequencies, they're not. They are just as directional as any other pressure wave radiating from it's source. Human perception makes location more difficult because of how the longer wavelengths interact with environment . . . Reducing the reverberant component. Subtly altering the reverberant component can, at times, be as reasonable approach to balancing some frequency (EQ) elements as twisting a knob.
In that sense, so in all those ways, then, no . . . No 'decent' mix without using tools associated with both 'time' (decay) and dynamics (attack) as well as amplitude and frequency.
