Short answer- we don't know without an additional metric of information that clues us into the performance level of the lifter
that day. Bar speed, RPE, or other metrics could be used to suss this out.
If the first workout was very easy for the lifter, e.g. he could have done 305 (or 310) x 5 x 3 sets on that day, yet only did 300- then he was
stronger on that workout as far as we can tell. This requires us agreeing that 3 sets of 5 reps is the display of strength, e.g. the "test". If we agree there, then another piece of qualitative data can be used to compare workouts. It may be important to do this to be sure that the appropriate stress is applied to the lifter on a workout (or series of workouts) as verified by the demonstrable adaptations.
Any stressor that does not produce useful fatigue and subsequent adaptation, as verified by successful improvement in the "test" or metrics that are being tracked (like a rep max or 3 sets of 5 in this case), is a programming inefficiency.
A workout can provide the wrong stress, too much or too little stress and the end result is reduced performance. For instance, 3 sets of 5 at 305 is not always more stressful than 300 for 3 sets of 5.
Again, performance has to have both qualitative and quantitative data points.
Interestingly, we use this in coaching folks all the time- novice or otherwise. If the bar speed looks faster during warm ups we'll make a bigger jump for work sets, maybe 10lbs. If it looks slower or the form is off, we make a smaller jump or maybe don't go up at all. All of this is in an effort to apply the right stress for the workout or series of workouts.