Video Explanation: Video
Please don’t judge the video too harshly. I’m a noob at making videos. 
Basically the theory is that the ability to score in the far zone doesn’t effect the rate at which that team scores points. This is because when you score stars in the far zone, you make it so that the other team descores more points when they throw those stars back over to your side.
I’d love to hear smarter people than me confirm or debunk this theory.
It’s really about when the clock hits 0:00… Then it really matters. 
In early season matches this may be a very good strategy, but it may not hold for the entire season. It is more about the time it takes for the opponent to react than which zone it gets placed in.
Getting elements to the far zone makes the other robot need to travel further to grab the element prior to getting it over the fence again. But that’s only assuming they spend their time in the near zone. However, if the opposing robot can send it from the far zone, then maybe this does not make that great a difference to those robots because they may be waiting for the elements to come their way.
Scattering the elements across the field would be like playing tennis by having a baseline game and a net game. It’s hard to cover both but there is something to be said about getting the opposing robots to collide with each other and not work efficiently together in their futile attempt to move the elements back to your side.
Putting the elements where the opposing robot needs to travel or turn around to get them makes them waste time. Putting it in the corners can also be a good place if the opposing robot is not adept at getting them out of there.
I completely agree in early season it does not matter but in your ideal alliance one of you will score from the fence and the other will score from the far zone in my opinion but both should be able to hang. Let alone have a bot that can do all 3 things proficiently. I will also say that the more variety of were and how you can score will help keep the odds on your side.
For the same average “rate of fire”, a robot that lands stars in the far zone will tend towards a higher score, since an opposing robot cannot instantaneously react and return those stars. If facing against a robot that must collect stars and then drive to the fence to score, a robot which can score from a distance will have an advantage by forcing the opponent to waste more time driving. Faster rate of fire than the opponent will amplify this effect further.
However, a dumper design that can deliver several stars at once could get an edge over a single shot rapid fire design by collecting and scoring a large load near the the end of the match such that the single shot robot cannot return them in time.
At lower levels I expect to see instances of high output robots that can overwhelm an opposing alliance almost by themselves. I suspect things will become very interesting at higher levels when all teams in a match have high outputs and/or long range capability, resulting in very close scores.
This strategy would make sense under the assumption that the game is scored at regular intervals, not at the end of the game. Sadly it’s the latter, which makes the end value of your game pieces the important factor. You’ve made it seem like the red robot is a dumper, and that the blue robot is a shooter, so I’ll run off of that.
The blue robot can shoot, so, assuming that it’s well designed, it can shoot from the near or far zone. Thus, it makes its cycles faster than the red robot who has to grab stars from the far zone, then proceed to bring them to the near zone, and over to the other near zone for scoring.
Of course there’s also the 2* factor on scoring value between near and far. You want to be able to score into the far zone - the factor is pretty heavy.
Taking this into an ideal scenario, this scenario would be, well, non-existent. The constant cycling is simply a matter of clutter. An ideal scenario would be hoarding the game pieces, then throwing them over close to the end, preventing de-score.
In summary, this constant de-scoring only matters to the extent that you keep up to rate with your opposition. After that, it matters in regard to which zone you’re shooting. Far zone robot will win in the end because, even if they end with same stars on opposite sides as the near zone robot, their (far zone robot) point value will be 2* that of the near zone robot.
It is important for teams to do the math to consider how much they have to outscore the opponents with near zone only pieces to beat alliances that score in both zones, with and without hanging robots on each side.
The short version for me, personally - scoring quickly and often is more important than scoring far. Try to score far, for sure, but not at the cost of scoring speed or cycle time.
I think autonomous might prove to be interesting in this regard.
Picking up and scoring objects already on the floor at the start of the match in autonomous will swing the point difference by 2-4 points, while scoring objects on the fence will only gain you 1-2 points. So you’d be better off picking up your own stars in autonomous.
shooting into the far zone makes it easier for dumpers
Lets go with your scenario of a 1v1 match between a blue robot that scores in the far zone vs a red robot the scores in the near zone. You are correct that during the middle of the match, the score will not change much since the point swings even out. However, what you are overlooking is that the blue robot gains a lead at the beginning of the match by scoring in the far zone and maintains that lead at a static level throughout the match.
What I mean by this is assuming the two robots move objects at equal rates, at the end of the match half of the objects will be scored for the red robot in the near zone and half the objects will be scored for the blue robot in the far zone. This gives blue twice the score of red, a blowout victory. Although the points were traded equally for the majority of the match, blue wins because the points scored at the end cannot be descored, and thus score in the zone they are in.
Red would need to have over 2/3 of the objects scored in the near zone to win in this example. Lets assume there are only 27 objects of equal value and ignore autonomous and hanging to make the math simpler. If the blue robot scores 1 star per 2 seconds in the far zone, this equates to 60 stars in a 2 minute match. The red robot needs to score 60+(2/3)*27=78 objects in the same amount of time. 120/78= 1.54 seconds per star. This is the rate the red robot will need to score objects in the near zone to keep up with the 2 second per star blue robot scoring in the far zone.
Please let me know if I screwed anything up.
The heart of the discussion, really, is “how true is this assumption”? If it is absolutely true, and a team cannot build a near zone only scorer that scores faster than the far zone scorer they could have built, then obviously the far zone scorer is preferred. The question really is, how much faster can you make a near zone scorer?
This answer varies based on the capabilities of the team making the decision. I’m sure the very best teams in the world can score just as fast in the far zone as the near zone, but maybe a middle of the road team could not figure this out. Better to be really really fast at the “worse” task than to be slow at the “better” task.
It’s also worth noting, in terms of point swings, a “their far to your near” scorer is changing the points delta by 3, which is theoretically the same as “their near to your far”, right? It’s more complicated than that, but still.
I addressed this question in my last post. The answer differs depending on the scoring rate of the two robots, but if a far zone scorer scores 1 star every 2 seconds, the near zone scorer needs to score 1 star every 1.5 seconds to beat it. The math is worked out in the last paragraph of my previous post. I don’t actually know how much faster a near scorer could be than a far scorer, but it would need to be this much faster to be a smart design choice.
Another factor to consider is scoring the far zone will only matter during the last part of the match, as only the points that do not get descored will count. If a robot was able to switch modes between scoring near quicker and far slower, it could score near for the first 3/4 of the match and switch to far in the last 1/4 which could be effective.
Does far zone matter?
Depends, skills ofc
In real matches, if you can only score in the near zone, you have to make sure you can score way faster than your opponent so by the end of the match, their far zone bonus won’t matter.