I have cycled and run with a HRM for approx the past 3 years. I only really use it for training purposes, but tend to wear it always out of habit. I would say though, despite having a fairly good understanding of several of the most popular training methods utilising HR, I did not find it particularly useful on its own, there is just too much going on much of the time.
Further, HR responds slowly so it is virtually useless for short high intensity efforts.
To answer the latter part of your question, what gains am I expecting from a HR monitor - if you mean performance gains directly from the meter, NONE, the HRM provides me with a measurement of how fast my heart is beating, which I then use as a proxy for how hard I am working, based on the fact that the heart responds to muscles demands for oxygen. I do however expect to gain a number that represents some combination of factors, one of which is how hard I am working.
You can get performance gains from knowing your heart rate and recording the data. You can analyse and record changes in heart rate whilst training the heart like you would a muscle to improve performance
Not quite, while your HR (max, resting and LT etc) may change over time, usually by very small amounts (RHR changes more evidently), you would not gauge fitness based on changes in HR. What you will find is that your performance (speed etc) increases for a given HR.
The problem with HR is that it is a metric that is not exclusively indicative of the effort you are putting into a physical activity and as such is one that should not be used as any form of authorative or sole measure. HR is affected by too many things, temperature, illness, stress etc to provide much information on its own.
A brief example of where HR training falls over: On Thursday, I set about my normal 2x20 min threshold training session, I started my 1st 20 min effort at ~295 Watts, this was a considerable effort, but also a slightly conservative start such that I wouldn't blow up within the 1st few minutes and could increase my output a little as the effort progressed (probably to 300W or there about). But, within a few minutes, I was finding even 295 watts to be quite tough, my HR was higher than usual, 92-93% MHR and I felt like I was really hanging on rather than sitting on a controlled but hard effort. I managed to hold on and average 292 Watts for this effort (lower than target, but still above threshold so there was still a training benefit).
Now suppose, I had been training with HR alone (so none of these power figures), I would have started the effort, observed my HR ramp up slowly over a few minutes (rather than the instantaneous measurement like the power readout), started to find the effort quite hard, looked at the HRM and seen 92-93% and WRONGLY figured I was actually going HARDER than I was aiming for and backed off to get down to ~87-90% MHR. This would have put me under my target output and I would not have been training as effectively. Something on that day, possibly poor sleep the night before, possibly fatigue from a previous training session, possibly the temperature, possibly hydration, possibly nutrition, possibly even the time of day I was training at had caused me to experience an elevated HR for the same effort as usual, what the cause was, I do not know. But what I do know is that with just the HR figure to go on, my training session would have been completed at a lower level and I would have been completely oblivious to this fact.
Without some other more objective data (a power meter) I would have been at the mercy of this uncertainty and would have essentially either wasted a training session by not going hard enough, or potentially done myself harm by continuing to train when there was something wrong (if my power had dropped by more than 5% below target I would have abandoned the session and rested but because I could maintain an output within this region, I continued, but it was very difficult).
None of this is to say training with HR is useless, it is not, it is better than training absolutelly blind, but there really are some pitfalls that you ought to be aware of.
PS.... you can not compare HR data between athletes, it is a completely individualistic measurement. It is not like W/kg which can essentially be used to compare athletes and predict the outcome of certain scenarios.
PSS... I wrote a lot, possibly too much, given the time of day, errors and nonsense are to be expected!