Which FRFR system should I buy?
"I have a modeling processor and want to run it through a FRFR amplifier. Which one is best for my needs? How many watts is enough? What else do I need to know?"
There's no good short answer to these questions. As with everything else musical, the short answer is always: "it depends..."
That's the bad news. The good news is that it's a lot easier to pick out a FRFR amp for a modeling processor than it is to pick out a guitar amp. When you run a modeling processor through a FRFR amplifier you get most of your tonal coloration from the processor itself. The function of the FRFR system is to make it possible for you (and your audience, whether it's the cat or a room full of adoring fans) to hear your processor.
Playing volume is the first thing you have to think about when you're scoping out a FRFR system. Are you going to be playing for yourself, perhaps with a prerecorded backing track, or will you be playing for an audience? Playing for yourself is the simple case: it's a question of how loud you you'll want to (or can reasonably) play. Typically this is limited by external factors such as family members and nearby neighbors.
If you'll be playing with a band then your desired volume is going to be determined by the other band members. If you can all agree to not play at a volume that'll sterilize low-flying birds and ground-burrowing rodents, you'll can get by with a lot less power. Also, if you'll be playing in live-music venues that supply a sound-reinforcement system and engineer (in other words, if the venue mics your instruments), then you'll need less volume (and power). If the venue happens to provide a good stage-monitor system with a separate mix for each musician, you shouldn't need to bring your own amplification.
Different genres of music have differing power demands, too. The more low-end your music demands, the more power you're going to need. Also, low-frequency reproduction demands larger speakers at higher volumes. This is a fundamental fact that you can't escape, Bose marketing notwithstanding. You can fool the listener into thinking that there's low end even if the speakers are too small to reproduce the lowest octave, but if you want to feel the thud of the lowest strings on your instrument you really can't cheat the laws of physics.
Of course, none of the above decisions are without consequence. More powerful systems with larger speakers have larger cabinets. They tend to weigh more and cost more, as well. You have to ask yourself: how much gear do you want to lug, and how much do you want to spend?
Unfortunately, there's no formula for any of this. More precisely there's no formula that any of us would find useful, as it would involve making measurements beyond the technical capacity of most of us. We have to fall back upon the one tool we do trust: our ears. Basically, I urge you to approach this the same way I urge musicians to approach all their other purchases: try to use your own gear along with the new gear you're considering. Play it at the volumes you'll be playing. If you can, rent the new gear for long enough to try it in your own environment(s). The room and the presence of other musicians can make a huge difference in your perception of the new gear. In short, play what you usually play through a bunch of FRFR systems in your price range. If a system sounds good and meets your requirements for volume, it'll probably be fine.
Probably... There's always a disclaimer, right? In this case there are several.
First of all, not every FRFR system will sound exactly the same when played at the same volume. Larger speakers and cabinets will tend to produce better low-end response. The midrange and high-frequency response characteristics will also differ in ways that may or may not be important to you. The differences aren't as important as they are with guitar cabinets, where the speaker and cabinet is expected to shape the sound. The differences are more like the differences among high-fidelity speakers: you may have a preference for one over another even though the differences aren't necessarily dramatic.
Secondly, more power typically means greater size and more weight. However, this is only true for comparable technologies. Recent advances in power amplification have given us amazing power-to-weight ratios. That's the good news. The bad news is that you can't always believe what you read in manufacturer's literature. Marketers all tend to play the "specsmanship" game, attempting to make their product seem more attractive by presenting exaggerated and unqualified figures for power and frequency response.
The fact that vendors tend to play games with power specs, combined with the fact that the generated sound is a function of both power and speaker efficiencty (which is usually not rated at all in an all-in-one FRFR system such as a keyboard amp or portable PA), makes "how much power do I need" a fairly useless question to ask or answer. On the other hand, FRFR systems are not like guitar amps in that they don't need to be "pushed hard" to sound good. In fact, FRFR systems tend to sound worse when pushed hard. Take that observation and combine it with the fact that clean solid-state power is relatively inexpensive and the advice becomes: once you find a couple of competing systems that sound good to you, buy as much power as you can afford within your price range.
Of course, there's a caveat to that advice as well. Power is going to cost you in terms of weight. Ten or fifteen pounds might not look like a lot on paper, but it can make a big difference when you need to carry it a couple hundred feet from your car to the venue. Of course, feel free to ignore weight if you have a road crew, unless you also have to pay for freight...
See, nothing's as simple as it seems...