Trying to Rotoscope And Have Low Qaulity, is it too low to Rotoscope?

Pulling a clean matte from that footage will take a lot of work. You face extreme color compression artifacts, motion blur, low contrast, low saturation, color bleed on edges, and a few more problems. It will take multiple layers with multiple effects and some hand Roto (masking) to generate a usable matte. 

Quality rotoscoping in Photoshop, even with great footage, is extremely difficult. If you only want to isolate one or two players, I suggest using Mocha AE to rotoscope the player. You can speed up the process with Translation (movement) tracking on the player, then use the Paper Doll method of animating multiple masks for each player. You'll get better results with less chatter than Photoshop's masking tools can produce on multiple frames. You might be able to pull a matte from one frame fairly quickly in Photoshop, but the edges on the consecutive frames will have almost no chance of matching, and you'll end up with a lot of edge chatter.

You will probably end up with many layers you will need to combine to create a usable matte. I do a fair amount of this type of Procedural Matte workflow, and it almost always takes combining four or five layers with different color, blur, and channel effects to create a clean matte. The goal is to create a clean black and white layer you can use as an Alpha source with Set Matte or a Luma Track Matte.

Edit: I'm working on a tutorial using Procedural Mattes created with the abovementioned techniques. Here's a rendered shot showing the before and after of 2 Osmo Mobile shots combined to replace the sky.

If the tutorial series is finished in the next few days, I'll post a link. It might help you in the future. 

This is nothing I would use Rotobrush on at all nor is it necessary to employ convoluted manual masking. All the info you need is there and can be extracted using channel operations and adjustments. It may just take multiple passes to get a good combined matte. E.g. the dark blue parts of the clothing can simply be extracted based on luminance and so can the reds and whites. Effects like Channel Combiner, Shift Channels, Levels and Curves are your friends. You just need to employ them strategically and play around with their order and settings, pre-compose a bunch of duplicates and then combien everything. The only manual masking you're likely going to need is the edges of the field and of course slight corrections on individual segments. It's still going to be some work, but definitelky better than trying to fix a botched RB or manually creating cutouts in PS.

Mylenium

I've moved this from the Using the Community forum (which is the forum for issues using the forums) to the AE forum so that proper help can be offered.