Was Aerial Assist Better than Ultimate Ascent?

[JesseK]

If despair.com came up with a Demotivator related to FIRST, it would involve bad partners vs good opponents during qualification rounds of Aerial Assist. I suppose it would snark on the single good offensive robot being double-teamed or sandwiched in between two opponents while the two mediocre partners were shut down by a single great opponent.

However, I have yet to coach a single FRC game quite like Aerial Assist. In a single match where both alliances came out with 200+ points match we played 4 different roles and had a great time doing it.

IMO, Ultimate Ascent was still the best combination of engineering challenge and inspirational game play. Plus there weren't so many beat up robots at the end of a Regional

[Pjtruslow]

Rebound Rumble was great. Ultimate Ascent was fantastic! aerial assist was good, but I absolutely loved the game last year and don't think you could easily beat Aerial assist. I want to see an off season event bringing back aerial assist, given that many teams still have their robots in functional condition, and it was such a great game.

[Orion.DeYoe]

Aerial Assist can't even dream of thinking about coming close to being better than Ultimate Ascent...

Originally Posted by Gregor

Up until division eliminations, there were still teams that could barely contribute. Yes that happens every year, but it wasn't nearly so devastating as this year. All other problems with this game aside, that was the killer.

I agree completely.

[BarryCurcio]

I thought Aerial Assist was better, there was much more teamwork involved with the actual game and one of the main objectives is to demonstrate teamwork and "Gracious Professionalism". I did see a previous comment questioning strategies but honestly depending on our alliance partner we were required to change our strategies on more than one occasion. I thought one active ball on the field was a great idea because if a robot had trouble getting it off of the field it was interesting to see how alliance partners helped.

[BrendanB]

Aerial Assist doesn't come close to Ultimate Ascent. The core of AA was good and had potential however combine the insane issues with penalties, bad game management FIRST put referees in, and the fact that sadly most teams just can't play the game each year it put AA as a bad year for teams who built good robots.

Sadly for many teams AA came down to a luck fest of who you were paired with (both good teams and even just teams who could get possession) and where your referees locked.

[DJB11]

Personally, I don't think it was. Just because Ultimate Ascent was more exciting, in my opinion. It's hard to put a label on it, because it's a matter of preference. Aerial Assist has cooperation and teamwork,while also providing more strategy options than Ultimate Ascent. It could go both ways.

[Siri]

Originally Posted by Kimmeh

What if instead of you directly helping your parners, you ask another team?
"Hey Awesome Team! I see that Struggling Team is having some problems with their random component. We play with them soon, and then you play with them later. Unfortunately, we're busy fixing our robot/another robot and don't have the manpower to fully help them. Do you have a few people who can join our person/people in helping them get going?

No one is necessarily asking you to do it all. However, by definition, veteran teams have more experience and connections than a rookie team. Not to mention it can be intimidating for a rookie to go up to a powerhouse team to ask for help. Or maybe they don't realize that there are other teams willing to help them. Serve as a middle-man and help them get help.

I'm all for doing this, and I try to both on a team and as an inspector. But honestly, most of the in-match problems I see aren't from mechanisms (except drivetrains, which I know some people manage to overhaul at events, but I've never seen much success). They're from inexperienced drivers or inexperienced/unresponsive coaches. It's not that they're not trying, but I've had several teams this season (in our relatively lucky schedule) that just wouldn't get to their assist or avoid fouls, no matter the pre- and in-match guidance I could give. That's what AA has really made salient to me, and preparing some of the lost drive teams I've seen to take on hard matches isn't an challenge that's conquerable during an event. Best I start earlier next time!

[AaronEllsworth]

From a spectator's perspective, and only comparing the years I have been involved which is just Ultimate Ascent and Aerial Assist, I liked AA much better. Keeping track of 2 balls was doable, countless flying frisbees was not. Add the fact that at least in Arizona and Las Vegas the automatic scoring system was often incorrect and you had to wait for the manual counting to know what really happened. In both games I thought the weighting of penalty points was excessive. Managing the big ball was clearly more difficult than many teams anticipated, but I really think the Aerial Assist was much funner to watch overall.

[Kyle Stewart]

Originally Posted by Vupa

I personally believe it was quite the opposite this year. The game was actually very unfriendly towards box robots because of several different reasons. The reality being that teams that had difficulty with building a robot also had difficulty maintaining said robot. Unfortunately these teams are often rookie teams or newer teams, meaning that they are also not used to the rules of the games and the nuances of competition. Multiple times in competition you would have to tell a team to not do anything or to not use their shooter or intake because they would "get in the way" or "cause a dead ball" perhaps even net some unfortunate fouls. And although this seems like a poor showing of Gracious Professionalism, it is the sad truth because as a team, you don't want rookies to lose you a match due to fouls. Thus actually putting the restrictions on the simpler robots and their ability to participate in the match.

This is why this game is worse for simpler bots. In Ultimate Ascent there is no such thing as telling your rookie teammate to stop playing in fear of fouls. You let them play defense and if they could score you would tell them to do their best. In Aerial Assist there is just one ball per alliance and thus it is key on who has possession, if your alliance member is unable to actually do anything with the ball, then you want to minimize your risk of them getting the ball. Thus reducing their role to perhaps even less then mediocre defense.

I think there are two distinct features to consider, the game design and the rules implementation. A little bird at CMPs told me that the GDC comes up with the game and hands it to a different committee to write the rules, and it was the latter that caused the vast majority of issues in AA. Poor rules implementations defining possessions, penalties, and assigning the point values kept AA from reaching its full potential, and in that regard UA is superior.

To me, AA is several fold better than UA in terms of its game design. It is the first FRC game that actually resembles a team sport, and is far more accessible to spectators because there's typically only two game pieces. The high level of interaction between robots made the games much more gripping than last year's solo-fest. While the game design was far from perfect, the real issue was that the rules were written in such a way that issues were almost inevitable.

As for teams that couldn't possess the ball, I'd like to share an anecdote. In LA, Team 687 had a drivetrain-bot that Team 294 provided a lawn chair for so that they could at least inbound the ball. 687, affectionately dubbed ChairBot, became a crowd favorite and was selected for eliminations. For Vegas, 687 had managed to add a functional inbounding mechanism. We selected them as our second pick because they were really well driven, could do quick inbounds, and were effective defenders. They performed well, sadly being damaged over the course of 6 brutal matches in the semi-finals (3 separate field faults -- one was due to the cameraman punching a ball heading to our human player). We lost the finals in large part due to their broken drive train, but we feel like they played extremely well prior to that event. I still think we made the right choice with 687, and it was just the field faults that put more wear on their drivetrain than it could handle.

Teams also got good at herding or pinning the ball to get assists. If they didn't know about that, it's the job of their stronger alliance partners to inform them of that and help them get those assist points. It's more work, but helps make drivetrain-bots feel included in the competition.

My point here is that the problem isn't with the drivetrain-bots -- the problem is that teams would rather blame their teammates than come up with creative solutions when partnered with drivetrain-bots or limited functionality bots. Teammates can and should use this opportunity to develop both their GP and improve their strategic prowess.

I understand the frustrations of Strength of Schedule and inconsistent penalties warping the outcome of games -- we had to deal with both in CMPs. However, these aren't really issues with AA per se. The problem is that a game that is fantastic in principle had a slightly flawed implementation. Warts and all, I still think it's a big improvement over UA, and FRC is heading in the right direction.

Now all we need is for someone to give FIRST an algorithm to auto-balance match schedules based on teams' previous performances...

[JackN]

I am intrigued by people who say this game has more strategic depth than Ultimate Ascent, mainly because I see the polar opposite when I look at the two games. The difference could come in definition of the word but lets think about this.

Aerial Assist had two viable strategies for alliances to pursue. These played out in the finals with the 1678/1114/1640 alliance vs the 254/469/2848 alliance. One alliance was simply making two passes and scoring to limit the danger of putting a ball in the air and having two inbounds. The other was making two passes and added a truss shot. Outside of those two options you had a one niche strategy of the one pass+truss+score. Now the in match actions of robots were very important, I will agree and each individual robot needed to act on their own and make decisions, but those decisions were seldom more than can I go pass or should I play defense.

Ultimate Ascent went in a very different way. Look at the four Einstein alliances from 2013. I would argue all four of them attempted to do very different things. The 33/469/1519 alliance attacked by using two full court shooters to flood the field with frisbees and then using their floor pickups to sweep up the missed shots, the 1241/1477/610 alliance just effectively ran cycles and played the odds that their opponents would miss more shots doing fancy things, the 1678/148/862 alliance had two defenders and one full court shooter and the 1640/303/3476 alliance attempted to do everything on the field with a 30 pt hang+FCS+cycles. Saying that the game lacked depth or that it was just watching teams running cycles is very reductionist.

I thought Aerial Assist was an ok game that had potential and good moments, but on the whole lacked the variety of some of the better games in FRC history. It was more hated on than it should be, but I would rank it in the bottom half of games that I have participated in (all of them since 2005).

[themccannman]

Originally Posted by JackN

I am intrigued by people who say this game has more strategic depth than Ultimate Ascent, mainly because I see the polar opposite when I look at the two games. The difference could come in definition of the word but lets think about this.

Aerial Assist had two viable strategies for alliances to pursue. These played out in the finals with the 1678/1114/1640 alliance vs the 254/469/2848 alliance. One alliance was simply making two passes and scoring to limit the danger of putting a ball in the air and having two inbounds. The other was making two passes and added a truss shot. Outside of those two options you had a one niche strategy of the one pass+truss+score. Now the in match actions of robots were very important, I will agree and each individual robot needed to act on their own and make decisions, but those decisions were seldom more than can I go pass or should I play defense.

Ultimate Ascent went in a very different way. Look at the four Einstein alliances from 2013. I would argue all four of them attempted to do very different things. The 33/469/1519 alliance attacked by using two full court shooters to flood the field with frisbees and then using their floor pickups to sweep up the missed shots, the 1241/1477/610 alliance just effectively ran cycles and played the odds that their opponents would miss more shots doing fancy things, the 1678/148/862 alliance had two defenders and one full court shooter and the 1640/303/3476 alliance attempted to do everything on the field with a 30 pt hang+FCS+cycles. Saying that the game lacked depth or that it was just watching teams running cycles is very reductionist.

I thought Aerial Assist was an ok game that had potential and good moments, but on the whole lacked the variety of some of the better games in FRC history. It was more hated on than it should be, but I would rank it in the bottom half of games that I have participated in (all of them since 2005).

As a member of 1678's scouting/strategy team I can vouch for this. Really the only decision we made in the einstein finals this year was where to put our inbounder, either at the start of our cycle, or after the truss shot. Seeing as we wanted to be able to get the triple assist no matter what while under defense that was too heavy for trussing it was the obvious choice to put 1640 at the start of out cycle so that we already had 2 assists as soon as the ball entered the field.

In 2013 we had much more to consider and we played many different strategies throughout each match. 862 was largely on defense in almost every match as their superior driving ability was most valuable there. However, they also played counter defense to protect 148 while they lined up for the FCS, and even ran cycles on occasion. 148 had to be able to both FCS and run cycles. 1678 had to play defense and counter defense along with 862 to slow down the opposition and to protect 148 lining up for the FCS and also be able to switch off to picking up discs off the floor to either starve our opponents, or to put in discs that 148 had missed. In 2013 each robot played many more unique roles. However that's not to say 2014 didn't have strategic depth, it just didn't have the amount of depth that 2013 had.

This doesn't even consider how many more options you had to consider in 2013 for alliance selection. This year was basically choosing an inbounder/defense, a midfielder, and a finisher that was either an open field or a fender shooter. In 2013 there were countless different options to consider when choosing alliance partners ranging from defense, to FCS, to climbing, to scoring colored discs, to cycler, to ground pick up, etc...

[JYang]

Originally Posted by Link07

Absolutely not

^This.

Originally Posted by DCA Fan

No.

^And also this.

[Vupa]

Originally Posted by Kyle Stewart

As for teams that couldn't possess the ball, I'd like to share an anecdote. In LA, Team 687 had a drivetrain-bot that Team 294 provided a lawn chair for so that they could at least inbound the ball. 687, affectionately dubbed ChairBot, became a crowd favorite and was selected for eliminations. For Vegas, 687 had managed to add a functional inbounding mechanism. We selected them as our second pick because they were really well driven, could do quick inbounds, and were effective defenders. They performed well, sadly being damaged over the course of 6 brutal matches in the semi-finals (3 separate field faults -- one was due to the cameraman punching a ball heading to our human player). We lost the finals in large part due to their broken drive train, but we feel like they played extremely well prior to that event. I still think we made the right choice with 687, and it was just the field faults that put more wear on their drivetrain than it could handle.

Teams also got good at herding or pinning the ball to get assists. If they didn't know about that, it's the job of their stronger alliance partners to inform them of that and help them get those assist points. It's more work, but helps make drivetrain-bots feel included in the competition.

The example given is a wonderful demonstration of a box robot being able to contribute to the game with a few modifications and some quick explanations. I concede that this is perhaps where AA shines in the way that no other game has, allowing box robots to really show off their potential. However, the game is also flawed simply in the way that it restricts the potential of teams that aren't in-bounders, but are not technically capable enough to be a high-level truss shot or finisher. This is the scenario in which you are telling your teammate to avoid picking up the ball and just play defense because you, as an alliance, fear a dead-ball or wasted time. We can see it as restricting the potential of a middle-tier team as their robot is realistically being told to not use the functions they spent weeks designing and building. This scenario is unfortunately repeated time and time again in qualification matches due to teams being unwilling to lose a match due to their alliance partners fumbling the ball. Perhaps an occurrence only seen at the regionals my team attended, it seems that the appearance of only one game piece per alliance has raised viewer-friendliness but at the same time also added restrictions on the other robots playing the game.

In the end, I still believe that Ultimate Ascent was better than Aerial Assist simply because of the fact that UA was much more versatile than AA. This can be seen in the different number of strategies that teams could go through involving different core fundamentals in robot design. (climbers, full court shooters, and cyclers, along with a host of designs in between). AA basically has one superior strategy of the triple assist in which robots all fell into and created designs with varying degrees of success for that one strategy. And although I admit that in the pure essence of AA, it was a good game due to its spectator-friendliness and attempts to promote teamwork, it falls flat in many aspects because of this lack of versatility.

[Lil' Lavery]

Perhaps there should be a distinction made between "strategy" and "tactics." While the "strategic" meta-game of highly competitive matches settled into 50-point cycles with trussing to the human player, that glosses over a lot of the finer detail tactical differences between alliances. The amount of robot-robot interaction, both in terms of assists/offense and defense was greater than any game in memory (even 2007 and 2009), and it led to a great deal of robot-specific tactical decisions, many of which evolved late in the season.

For instance, look at how assists were gathered. While the default method was one robot would pick up the ball and spit it out on the ground towards another, there was plenty of variance from that. "Kiss passing" was popular from essentially the beginning of the season, but not every robot could accomplish that easily (especially if your intake had to stick out dramatically to accumulate or release a ball). A number of robots had specific methods for passing to each other, like arm bots dropping a ball directly into an open hopper or what amounted to hand-off between bots with comptabale gripping mechanisms. Late in the season a new method emerged; consisting of quickly passing the ball back to the human player immediately after an inbound, and having that human player throw it to a second robot. This was used plenty during qualifications at MAR champs (as far as I can tell 1089 was the first team to use this), but really gained popularity in St. Louis, to the point where two different Einstein alliances (2590 with Archimedes and 973 with Galileo) used this method to rapidly register an assist without much risk of losing a ball.

You can find similar distinctions defensively, as well. 1712 developed different methods of harassing various inbounders, based on their machine's design. If a team didn't establish firm control of a ball during an inbound, we knew that a well-timed impact could knock the ball loose before they registered a possession. Others we opted to keep away from the inbounding position, since it was difficult to load them at range. Team 2590 could use their arm to disrupt pass-through inbounding and kiss pass attempts. Team 118 used their height to disrupt 254 from being able to score. 1625's swerve (tank?) and drive team were able to punish 33 for their drivetrain selection during the Archimedes finals.

There are plenty of examples of robot-specific match-up that opened new tactical options in this game, plenty more than I've seen in most recent games. Ultimate Ascent was pretty much limited to blocking lower release point full court shooters (or a rare attempt at blocking pyramid shooters), otherwise was limited to pushing matches and parallel play. You didn't really tailor your tactical decisions based on machine characteristics, beyond how many discs a team could score.

[Cam877]

This is a very controversial question everywhere in FIRST right now, and I have to say Aerial Assist really stole my heart by the end of the season. In the beginning, I wasn't too sure, for my team had a rough first event, with the qualifications filled with non-working robots and our elims campaign hindered by fouls. However, as the season went on, and I went to MAR Championships and then FIRST Champs, I realized that this game is astronomically better than ultimate ascent. As the competition got better and the strategies got deeper, the elimination matches and qualification matches alike became more and more exciting. The excitement of the elimination matches, however, simply cannot be matched by any previous game. After last year, I said to myself, "How could FIRST ever hope to make a better game than that?" And this year, I was blown off my feet by the sheer greatness of this game. And now, I find myself in the same position again, wondering how FIRST will be able to come up with a game even close to the greatness of aerial assist. And judging by FIRST's game making skills of late, I have no doubt that I will be pleasantly surprised again come next year.

AA > UA!!!