Here are some quick tips to help you gain enough motivation to finish your games. Most of them are obvious, but sometimes writing an article like this is motivating in itself .

Motivational Rush
You get motivated when you have something you really, really want to work on. You get even more motivation if you succeed at something. You get demotivated when you fail at doing something. And if you haven't worked on something in a long time, your motivation factor tends to drop by time. The more projects you have, the more parts you have to divide your motivations to. In simpler terms:

M is motivation
t is the amount of time you've worked on project

Idea quality = M
(Success)t = M
(Failure)t = -M
CurrentM = (MaxM)e^(-Xt), where X is your personal laziness factor
M = TotalM/(no of projects)


From this, we can come with the following conclusions:
1. Keep the amount of projects you have to a minimum
2. Jot down an idea when you get one, or you'll lose the free motivational benefits that come with it. Alternatively, don't jot it down, don't work on it, but keep it around so that you'll get the motivational benefits each time you get the idea.
3. Minimize the time you'll be spending on your game. A game stretched out over 2 years instead of 2 months can do massive harm to motivation.
4. If you're going to work on something particularly difficult, make sure it's possible and do it till you succeed. The motivational benefits of finishing an extremely tough engine is unmatched, but failing to finish that extremely tough engine could demotivate you into abandoning your project.


Minimizing the time it takes
You'll never have enough time to finish. One of Murphy's Laws states that the time you have available will fill the amount of time you need. In other words, no matter how much time you'll get, you'll never find enough.

A nice trick is to use Murphy's Law to your advantage. Instead of trying to reduce the time you need, you should reduce the time you have available. Give yourself a deadlline. A game that you have forever to finish WILL take forever to finish. A game that you plan to finish in a month will be finished in a month, as long as you truly try to finish it by then.


Building up motivation
As every success, every idea builds up a bit of motivation, make sure you build up as many small successes as you can. Do the little bits first. Every successful piece of the puzzle you place will give you small adrenaline rush. Every tiny problem you run into will breed more ideas. Once you've built up some strong inner motivation, you can get working on the big, tough engines.

Sometimes, even when you're not motivated enough to finish a tough physics engine or combat engine or something, it's best to do all the simple parts in it first. Get your plans on them done, create arrays, data storage, encryption, whatever simple things you need. Motivational boosts can last surprisingly short, maybe an hour or so, and you want to spend all that time thinking, not doing menial labor.


Starting it all
Finally, the hardest part is to actually begin. You start off with almost no ideas, no successes, nothing. Just a blank white frame. This could be the beginning of the game, a level, a new engine, new sprites, artwork, music, etc. Everything has its beginning.

The best way to start off would be to use the time method. Force yourself to do a few sprites, a few objects, a couple of event groups in say... a day. It works a lot easier that way.