That option is merely going to play in how big is the Fitbit adjustment at the start and end of the day.
If it helps with planning what you are going to eat - might be useful to pick the realistic one.
I'm betting you have a pretty good idea of where eating level should be anyway, and probably not needed until end of day to help get to eating goal.

The closer MFP activity level matches actual activity level - the less MFP has to correct itself using the Fitbit stats.

At the end of the day both choices would result in same amount of calories for eating goal though.

So your 750 deficit if followed is probably just barely reasonable for amount left to lose.

But you have made it a lot more it sounds like by missing your eating goal.

Your body isn't going to be happy eating upwards of 50% less than what it could burn, and it'll adjust so you burn less and aren't that extreme to what you eat.

Extreme deficit is going to stress body out and slow it down, also cause muscle mass loss. You still have a deficit - just not what it could be - hence the slower weight loss.

I'd suggest trying to reach your goal better, especially introducing more exercise.
You may even need to go to maintenance mode for a few weeks to let it recover and speed up to full burn - right now Fitbit's estimate is on healthy average body with your stats - your's may not be entirely healthy at this point and it's overestimating your burn.

The other difference between choice of activity level - is the inflated calorie burn introduced in the evening when you finally plop down and then sleep until midnight - when you are obviously not being Active or whatever level you pick.
Hence the suggestion to pick sedentary which minimizes that effect.

The FAQ in the stickies - 2nd half, discusses this with math to see what happens, as well as what MFP is doing with Fitbit data.

What heybales was saying was that you may be trying to lose weight too fast, causing your body to burn muscle and slow down its metabolism. (The rule I've learned says that a deficit of 750 calories a day is only for people who need to lose more than 50 pounds.) As a result, your Fitbit may be overestimating how many calories you burn. In order to correct the problem, you might need to stop trying to lose weight for a while - increase your food to a maintenance level and at the same time, increase your exercise in order to regain some of the muscle mass that you've lost. (To be honest, it sounds like you're pretty active at work already, so you might not to add that much exercise, just stop stressing your body by not feeding it adequately.)

To figure out if your Fitbit is accurate for you, take a month where you log your food extremely carefully - weigh every single bite you take to the gram and log it. At the end of 30 days, you can use numbers on the intake vs. burn graph on your Fitbit profile page to compute your expected weight change and compare it to your actual weight change. The difference will tell you how many calories per day (on average) your Fitbit is off by.

I also wouldn't call losing 3 pounds a month "extremely slow" when you've got as little to lose as you do. At most, you should be trying to lose 4 pounds a month (1 lb a week), perhaps even just 2 pounds a month. Maybe what really needs to be adjusted is your expectations as to what a reasonable rate of loss is.

You'll have to give it more than a couple weeks.

As a woman your metabolism literally changes through the month, pick the wrong couple of weeks for anything weight related you can have bad looking results.

One effect you should immediately get is more stored glycogen in your muscles, along with the attached water.
This will immediately increase your LBM, your metabolism, and your scale weight - but not fat weight.

So reminder not to be bound to the scale which has no idea if you've lost muscle or gained water, or it's fat weight.

This speedup will take more than a couple weeks.

If you were going to start any strength training - this would be the time.