What are the qualities of a good FRC game?

I know everyone is wrapped up in championship mode right now so I do apologize for posting a thread slightly off topic, but I’ve been thinking of a question that I would really appreciate the wide variety of Chief Delphi’s opinions on.

I love the concept of game design, and I’m wondering what everyone thinks are the qualities of a good FRC game. I think every game in the past few years has had its share of unique qualities that positively affected the game experience, and it would be really interesting to see what other people think about this topic as well.

Linear scoring

1 Like

A proper FRC game can be enjoyed at any level of play, be it regional or championship. In addition to being enjoyable at whatever level of play, a proper FRC game should also be entertaining to any audience without any need to explain or justify the look of the field, game objects, decoration, etc.

That the use of strategy is involved, and can dictate the outcome of a match, meaning you must know the game and design a good bot to be successful.

  • Drive teams can see the robots on the field
  • Spectators can see what’s going on
  • People watching online can see what’s going on

I like a game where alliances can benefit from collaborating between partners, but elite teams can still carry. The 2012 FRC game is a great example of this (ball feeding was a legit strategy, but by no means necessary).

Fun and interesting for students to compete in.

I think it important to have simple challenges that low resource teams can do. I really like the gear aspect this year because it was something that anyone can do.
And I think it is equally as important to have a more difficult thing that low resource teams can attempt, and the high resource teams can work to perfect. High goals last year and this year are a good example of that.

2014 was a great game yet didn’t have these, but it was a simple enough task that everyone could attempt but was tough to master and execute at the level of the 254’s of the world

AND simple to understand scoring. As soon as you have to describe the game to someone, and the phrase, “…but if you do Y first, then the points for X are…”, you’ve lost most people.

Multiple levels of tasks, all of which are easy to understand, with progressively larger points rewards. That’s what it takes. (ahem… VEX 2017/2018 game)

I would also add that a good game has to have an exciting, teamwork ending.

Ah, finally, something college prepared me for! I have a Comp Sci degree with a concentration in Computer Game Design.

Linear scoring is important for stats, but it isn’t a huge requirement, in fact, nonlinear scoring can often create more tension and require more strategic thinking. Which is cool from both an audience and a team perspective.

So what makes a good game? Balanced scoring options. This is a location where FIRST has been REALLY bad. (See Fuel)

Other factors:

Drivetrain only scoring option - Having an option for scoring that can contribute to alliance scores both lets lower resource and rookie teams contribute but it also cuts down on knock down drag out defense.

No Coopertition type task - tasks that are removed in playoffs and only serve to inflate rankings really contribute to bad alliance selection.

Ranking System not solely based on your schedule - 2016 really let the cream rise to the top, 2017 was decent too.

If I had to rank these:

Balanced Scoring
Drivetrain Scoring Option
No Coopertition Tasks
Ranking System

So then your favorite game should be 2008, right?

Accessible to all teams, regardless of ability.
A balance of judgement and non-judgement calls for fouls and scoring
Corollary, more automated scoring allows better judgement calls, but has more chance to screw teams over.*
Involves the entire alliance: One team should NOT be able to win or lose a match by themselves, regardless of how much or little their alliance partners contribute.
Balanced penalties for the objective of the penalty. **

*For a given value of “screw teams over”. If the automation fails… what’s the backup??? Boulder counting in 2016 wouldn’t have been too bad, and gear counting this year, but fuel counting??? NOT IT!
**This one takes some explaining. Basically, if the penalty is there for safety of personnel, a card is appropriate. Integrity of the game, also a card. Preventing certain strategies should be fouls of some X points, while preventing minor “you messed up” moves would be fouls of some Y points where Y<X, and the values of X and Y should be proportional to the number of points expected to be earned as a result of the moves or points. If the penalty is there just to be there, then why is there a penalty there again?

The most important quality for an FRC game is variety of types of robots. If a game can support multiple robot types, then it implies that objectives are usually fairly balanced.

Many people have already hit a lot of the important ones. One I would like to add is that FRC games are rarely games in the sense that your action affects your opponent. Even some of the most highly regarded games like Aerial Assist or Rebound Rumble were pretty much point scoring races that happened to be played at the same time. Sure, a good defensive bot could sometimes make a difference but this was honestly minimal. I think 2010 did a good job of making it an actual game. There were no purely no-go zones, so there was always a team defending the goals. Furthermore, with a limited supply of balls (and no new game pieces entering the field), a very limited carrying capacity, passing balls between teammates (or keeping them away from the opponent) was actually organically important unlike 2014 where it was contrived.

Of course, true “game” nature does not make a game automatically good. 2009 has a true game in this sense and was a bad game for other reasons. But I think that having more interplay between the two alliances rather than scoring races makes the game better.

2010 was the best first “game” ever

I like games that create a lot of robot diversity.

I like games that have an “end game” that is important but not overpowering.

I would like to see a game that is similar to vex where autonomous plays out like its own little match with a point bonus awarded.

I like a game that’s easy enough to explain to grandma.

I like a game that is easy to officiate and easy to avoid penalties.

  • It should present a real strategic problem to analyze - there should ideally be a significant number of viable strategies for how to play the game.
  • It should provide engineering challenge - teams should not be able to pull out their robot from four years ago, tweak a couple of wheel sizes, and be able to play this game well. Ideally, there will be several challenges such that each one is not terribly difficult, but putting all of them in one robot is.
  • The game should be easy to explain to family members, and for them to understand why the winning alliance won each match. Not necessarily linear scoring, but sensible scoring.
  • The game should be exciting - all the way through. A high autonomous score should be a big step up, but not a crushing blow to the opposing alliance. Relatively high-scoring endgame tasks give alliances a chance to bounce back (though 2017 was probably excessive).
  • Human player role should be supportive rather than scoring. 2017 gear/rotor scoring got this exactly backwards.
  • Rookie and low-resource teams should be able to reasonably contribute to an alliance. Ideally, second picks will be made primarily based on the robot and driver that the teams brought with them, not soley the fact that they are underweight enough to accept cheesecake.

Given that so many teams do rely on OPR, having linear scoring is kind of important. I think the rotor to gear OPR transform is super neat, but its still lacking for a lot of practical reasons. I suspect elimination matches are better and more deserving teams make the big dance when scouting is easier.

I think minimizing penalties and cards is important. When fouls and yellow cards are being handed out like candy, something is wrong. Half the tournament should not have yellow cards – I think that sends the wrong message.

A big finish is important. The huge number of climbing robots is a big plus for this year, and I’m still not really sure why it happened. Maybe we just got lucky that Velcro climbers were really easy, like indexing frisbees was really easy in 2012. In other years climbing has been really exciting, but we haven’t seen nearly so many robots do it successfully.

I think human & robot interaction can make the games a lot more exciting to the audience. I was very disappointed when it became clear that robots would not be vomiting cascades of wiffle balls into the boilers at most events, because robots shooting lots of wiffle balls looks awesome. BUT, the audience participation in Steamworks was still really high, because everyone was empathizing with the pilots. There were several opportunities for heartbreak, and the audience was following each gear along its ascent. Properly balanced, I think the symbiotic robot to human mechanism works really well (2004, 2017). The human as an independent scoring mechanism (2009), worked a lot less well.

The tiered nature of Steamworks is pretty neat, but I think impossible to balance across FRC. When it worked (Einstein SF 15, for example), it was so cool. But for most events, it didn’t really work and it was all rotors, with the differentiator being mistakes or fouls.

I think the field should be more open, but I think the limited visibility as a form of protection worked pretty well. There wasn’t a lot of defense at the pegs, at least in part because it was so hard to see. I don’t like the protected zone penalties because they can sometimes get out of hand, so I hope FIRST considers using that mechanic again.

+1 for robot diversity. My rookie year (Ultimate Ascent), the most memorable thing was the fact that there were so many ways to play the game, and that each robot was so visibly different. Triangular shaped corner-climbers, tall FCS, lever arm ground pickups, every robot just seemed different. Even Einstein had a ton of visible robot diversity. 2016 did this kinda well with short vs. tall, and 2017 did this poorly with the volume constraint.
The easier it is to recognize different engineering choices and outcomes, the better.

The audience should also be able to tell the score by looking at the field itself. One of my favorite parts of 2017 was being able to see the number of gears and start biting your nails when you knew there was only 1 gear left before the fourth rotor. I feel like FIRST is starting to take this into account, with their abundance of lights signaling scores on the field.

Tasks in series, not in parallel. As cool as it is to have 6 robots all do everything for 2 minutes, it is confusing. Having a scoring system that cannot be done simultaneously by all robots is far easier to watch, and is how all real team sports work. I think this is what made 2014 so great.

I’ve only been around since 2014, but there have been qualities in every game since then that contribute to making a good FRC game.

2014

  • Simple game to understand/watch
  • Meaningful teamwork/collaboration

2015

  • Interesting engineering challenge
  • Near perfect qualification ranking sort

2016

  • Optional robot size constraint
  • 2 scorers + D vs 3 scorers very well balanced

2017

  • Perfect human player interaction (pilots) with robots
  • Exciting end game that can swing the match