Week 1 Impressions of Rack 'N Roll

[platypus]

After watching the eliminations, I'll add to what I listed yesterday:

1. The ideal configuration is two scoring bots and one lift bot, but I could have told you that 1 minute into strategy discussions at the beginning of build season. Yet, this simple statement makes a huge assumption: the lift bot must be reliable. At GSR, there was one truly reliable lift bot, and it was a simple bot that let out a platform at one side and a gradual ramp on the other, with no manipulator. It was second-round picked by the 1st seeded team, and was climbed for 60 points every round.

2. This is a HIGHLY proactive (offensive) game. If you can score, you should never defend. Simply put, the proactive plan of doubling your score by adding to the row that you're completing tends to be more effective than preventing a single opponent from doubling his score for 20 seconds. Use PROACTIVE DEFENSE, by scoring in places that prevent your opponents from scoring.

3. The ideal basic strategy is to start by placing ringers on two side spider arms to set up a row of five. This prevents the opposing alliance from ever having a chance to make something big of the bottom row, which is the easiest to score on by leaps and bounds. It's proactive defense.

4. If in the rare case you have a reliable lift bot on your alliance, audible your typical strategy for the following:
-Use your scorers to proactively defend by placing ringers in places that stop your opponents from scoring a row longer than 3.
-Have your lift bot set up way in advance (25-30 seconds left in the game) so that your opponents cannot possibly stop him from getting to the endzone in time.
-At about 15 seconds left, you should have less points on the rack, but neither score will be even close to 30 points. Send both your robots to be lifted. If even one is, you will win.
This strategy was used over and over again by the winning alliance. Their opponents really never had a chance.

5. Some people thought that spoiler use might pick up on day 2... they were wrong. Proactive defense is the way to go. Spoilers are not proactive, and teams' abilities to place them suffer from the instant outcry from the announcer that, "OMG, THEY'RE GOING FOR THE SPOILER!!!!!!!!!"

6. There is much debate on this thread as to whether or not this is an exciting game. My conclusion: it really depends how deep you personally analyze these games. The general audience loves this game, as, unlike last year, it's easy to understand exactly what's going on, and it's very fast-paced (because proactivity is the key).
As a analyst myself, I thought this game was boring. The games were often all the same, where one team got control of the majority of the bottom row and it was all over with 30-45 seconds left on the clock. The only question mark was whether or not a lift bot would set up, and whether the other robots were capable of climbing it. In most games, there wasn't even a lift bot and most lift bots were not reliable (thus if you know the lift bots at all from scouting, you already know the outcome 9 times out of 10). If you see past the fast-pace, the game is deceptively shallow. I really hope that this changes as the season moves forward...

[Bongle]

Originally Posted by platypus

If you see past the fast-pace, the game is deceptively shallow. I really hope that this changes as the season moves forward...

I haven't watched any of the video yet, but do you think that as regionals move on, teams will start heavily defending against lift bots? As you said, they are essentially the game-deciders, so blocking or trapping your opponent's lift ability may be worth it. If you sacrifice one lifted robot to stop their lift robot from ever lifting, then you're still doing better than your opponents.

[Mike Starke]

Let's just say I'm super excited for FLR this weekend!

[Joe G.]

From what I saw at BAE today, this game is one of the best I've seen, particularly as a spectator game. Some of the the good things about Rack and Roll:

-Huge variaty of designs (arms, elevators, ramps, lifters, pure defensive
robots, and combinations.) Aim High 80% of the robots were pitching
machines on wheels, and Triple Play there was only one real task for
robots to do , so all the designs were kind of similar

-Game can change at any moment. One of the things I did not like about
Aim High was that 2/3rds of the time, the winner of autonomous would win
the match. This put a fair amount of teams at a disadvantage if they, for
whatever reason, did not have experienced programmers avalible, even if
they had a very good robot mechanichally. With Rack and Roll, the
outcome can change with one ringer at any moment, and this will
probably be even more the case as the season progresses and spoilers
become more of a factor.

-Lots of defensive opportunities. Although not quite as important as in Aim
High, this game encourages defence. A fair amount of teams have had the
strategy of repeatedly ramming the rack, which is quite effective against
all but the most experienced drivers. Opportunities for a team to score big
with one maneuver occur quite often (adding to a large row in a close
game, ramps, spoilers,) and for an alliance to do well, they must be able
to block at least some of these.

-Teamwork between alliance partners is required. Since defence is so
crtical, counterdefence is also quite important. Also, there have been
several instances of teams helping other teams up ramps, teams trying to
tip teams that have been tipped failing to climbing a ramp, etc. Along with
this, alliance partner selection for the finals is much more complicated, as
teams have to consider the two distinct elements of the game, and how
the three robots will work together to get the 60 points in the endgame.

-Spectatorwise, the game is great. No matter who you want to win a
match, you always seem to find yourself rooting for teams to score in
autonomous, just because it is so impressive. The action around the rack
it quite intense, and as mentioned above, the score can change at any
moment. If needed, teams often put on great defensive displays, and the
ramp climbing at the end is as exciting of an endgame task as the
endgame was in 03/04.

That being said, there are also some definate negative aspects of the game, but as others have said, a fair amount of these will probably improve as the season goes on and the rack gets filled up more.

1. Spoilers are not that useful. True, it is almost always more advantageous to score a ringer instead, but as the season progresses and the rack becomes more filled, expect spoilers to play a more important role. So far, most scoring has only been on one level per match, but in a situation in which both large horizontal and vertical rows have been created, a situation that will start happening more and more often, a spoiler placed in the middle will reduce scores by large amounts. Also, the lack of spoiled spiders was not so much due to them being completely ignored, but due to extreme defense played on teams possessing a spoiler, since they were usually only brought into play when an alliance had a 5-6+ row

2. Keepers do not give much of an advantage. I believe that as the season goes on, the importance of keepers will increace dramatically. As more of the rack gets filled, keepers will be used strategically as the middle of large clumps. This will make clumps like these harder to spoil. Still, keepers could use more incentive, like making them multiply rows they are in by something, or an additional 10 points or something just for having it on the rack, regardless of its placement in rows.

3. The top row is ignored. As teams become more practiced and used to the game, collums will become more important, and the large horizontal rows we have seen so far less so. Therefore, an ability to score up high will become a desired ability as the season goes on.

1. A robot should really need to do more than unfold (mind you, i love the unfolding robots and i would build one given the chance now. Imagine a robot that unfolded across the entire home zone... it wouldn't even need drive wheels!!) Even the best ramp bot should need to play some defense or do something besides open up.

I really didn't see many ramp robots doing this at BAE today. (may have been different at other regionals though) Most ramp robots would ram the rack/push other robots for the first minute, and then go and unfold with about 45 seconds left

However, one problem that not many have mentioned is that a human player can inadvertantly shut down their own robot scoring wise. Several times, a ringer thrown by a human player got wrapped around some long, tall part of the robot, dissallowing it from picking up other ringers. Unfortunately, I can't really see any way to work around this problem other than for human players to be more careful. Also, the "random" alliance issue really needs to be fixed.

[Joe Matt]

Originally Posted by platypus

If you see past the fast-pace, the game is deceptively shallow. I really hope that this changes as the season moves forward...

I LOVED your analysis of the game, but I find a big flaw with this statement. You're assuming a few major things with what you say...

1.) You can ONLY score one ringer on each spider leg. This is not true, anyone can go over what you put, but ONLY ONCE. Two can fit on. This is a proactive way to negate a large score. This adds a MASSIVE amount of new strategy to the game that only boggles the mind. Even though it's harder to see on the other side of the field, any scoring on the sides can be well seen and changed.

2.) You igonre the ramp bot quite a bit and fail to realize it's potential to act on defense during the match. If the robot has a good latch system and drive train, some pushing and ringer moving would give the other alliance enough of a headache to gain a small advantage.

3.) You fail to address the unpredicable nature of the tubes. These aren't sure fire scoring objects like balls, they are like bins, very unpredictable.

4.) The rack multiplication goes up and down along with side to side. A sqauare of 4 would get you the same amount of points as 4 on the bottom, but would be harder to break for the opposing alliance due to visibility.

I see this game as lacking certain aspects of previous year's games, but not to the point of being shallow. There's a lot there.

Oh, and I disagree that last year's game was hard to understand, by far, it was one of the easiest for new people to understand (shoot balls in the goals for 3 points, side goals for 1, 4 game periods, ramp at end) compared to other years games (balls fall, humans must shoot, can cap, only counts in colored goals, bots can hang, etc.)

Here's some things I agree on...

Spoilers- pointless at the current time, I think their inclusion was a test by FIRST, much as the inclusing of having robots interact with other robots to score (aka the ramp).

Keepers- they seem to lack any significant value during the qualifications besides making your team stand out, as we progress to The 4 and beyond (aka the 4 Championship fields and Einstein) they will be priceless, especially if they can score them on the sides of the rack, not just the part facing them.

Auton- will win games just like the ramps. If you get those 3 up there, that's just 2 more to put up till you negate a robot on their side. JUST TWO. One more would negate the max 60 points they can get by raising their robot.

[dbell]

At GSR, there was one truly reliable lift bot, and it was a simple bot that let out a platform at one side and a gradual ramp on the other, with no manipulator. It was second-round picked by the 1st seeded team, and was climbed for 60 points every round.

That is 1824 that you are talking about. 172 (the pink one) was also a very reliable lifter. there may be others too.

If you can score, you should never defend

No. We (1519) could have scored 2 or so tubes, but, with defence we were able to stop the aliance from scoring much more than we could have scored.


My thoughts on the game as a whole are this:
(this is my opinion, and I havn't seen a lot of rouds because I've been in the pits a lot. )

1. I would have liked to see more importance for the autonomous mode. The keepers didn't make muck difference. this auto mode reminds me of 2005.

2. spoilers sound like a good idea to lower the opposing the other teams score, but.... And often it would make a lot of sense to do this instead of adding another tube to your own row, but, the other team will probably stop that from happening with defence.

3. the human player involvment is good. once I saw a HP throw a tube onto a spider .

4. ramping can ba a huge thing or it might not. it depends a lot on the teams in the match. sometimes when each alliance has a few tubes and about the same number of points the ramp can make all the difference in the world. or sometimes 2 very good tube scoring bots can get a row of 6 or more and the ramp won't help. i think that this make the game interesting.

5. the bottom spider is too important. (my opinion)

just my thoughts

[VEN]

I think that there is too much defensive play this year and less that involves completing a certain task. Also, with the creation of the game a choice was given to create a ramp bot or a bot that scores and as it turns out many took the path of the ramp bot cause the game play to be violent and sometimes pointless.

Overall, from what I saw watching the New Jersey webcast, the game IS exciting and there are many sweet robots out there

[wilsonmw04]

Originally Posted by VEN

Also, with the creation of the game a choice was given to create a ramp bot or a bot that scores and as it turns out many took the path of the ramp bot cause the game play to be violent and sometimes pointless.

Exactly what i was thinking Ven.

[Andrew Blair]

Originally Posted by Joe Matt

I LOVED your analysis of the game, but I find a big flaw with this statement. You're assuming a few major things with what you say...

1.) You can ONLY score one ringer on each spider leg....

Um, not true. Once you have placed a ringer onto a spider leg, that leg cannot be scored upon by the opposing alliance. It's value can be negated by the placing of a spoiler, but once it has been scored by the other team, you can't ever get it back for your alliance.

Please refer to rule <G15>:

<G15> Double RINGERS - a RINGER can not be placed on a SPIDER LEG on which a RINGER is already HANGING. If a ROBOT HANGS a RINGER on a SPIDER LEG that already has a HANGING RINGER from the opposing ALLIANCE, the second RINGER will be ignored when calculating the match score. If a ROBOT HANGS a RINGER on a SPIDER LEG that already has a HANGING RINGER from their own ALLIANCE (for example, to prevent a SPOILER from being placed on the SPIDER LEG), the two RINGERS will cancel each other and both will be ignored when determining if the SPIDER LEG is SCORED.

[Hieb]

Originally Posted by Joe Matt

I LOVED your analysis of the game, but I find a big flaw with this statement. You're assuming a few major things with what you say...

1.) You can ONLY score one ringer on each spider leg. This is not true, anyone can go over what you put, but ONLY ONCE. Two can fit on. This is a proactive way to negate a large score. This adds a MASSIVE amount of new strategy to the game that only boggles the mind. Even though it's harder to see on the other side of the field, any scoring on the sides can be well seen and changed.

You better check <G15> on that. Double ringers are either ignored or negate your own score.



A few other issues:

1. I don't entirely agree that the bottom spider is the most important. While it may require the least lifting, it is also the easiest to defend regardless of the size of your 'bot. A long arm, on the other hand, can reach over the top of a defending 'bot and still score.

2. Depending on the alliance, a good defense ("proactive" or not) can be very important. If the opposing 'bot can easily score 4 or 5 ringers undefended and I can hold him to just 1 or 2, I've made a big difference in my alliances chances to win. Similarly, if I can "defend" my alliances scoring 'bot by not allowing the opposing defense 'bot to interfere, I've done the same. As far as not being meeting the goals of the game--perhaps not the way you want to play it, but that is something I love about FIRST. Every year I see solutions to the problem that I never even imagined. We specifically chose a gearing that would allow us to to do well in pushing matches (I can't remember ever being out-pushed) and would also allow us to power our way up a ramp. This turned out to be very important, as we fried 3 banebots motors on our arm before switching to pneumatics and then finally abandoning offense.

3. How do you define "agressive defense"? Constant pushing and shoving to keep a team from scoring seems to be very much in the intent of the game, especially with the bumpers that protect the bot while giving a weight advantage over teams that chose not to use them. High speed ramming and similar acts are spelled out in the rules and punished accordingly (sorry Channel Cats--our all out autonomous dash across the field in high gear was a programming error--and the penalty also cost us the match).

I think overall there are many more aspects to the game than I originally thought, and some of the things I though would be important turned out not to be. I really wish we had another opportunity at it, but now I'll just have to sit back, get some sleep, and then see how the rest of you handle this game.

Good luck to all.

[Andrew Blair]

Something to note: the regionals seem to display greatly different competitiveness. I had focused mostly on VCU, but after watching a few BAE matches, the game is much faster paced and exciting in Manchester. I think the teams at the regionals will control the play for awhile.

[Neo3One3]

Originally Posted by d.courtney

4. The last thing that killed the game is the "random" schedules. This takes half of what FIRST is and throws it out the door, instead of having it so that no team is against any other team, playing with one match and playing against the next, we find many teams playing the same team time and time again. It also punishes the teams on the receiving end of a "stacked" side when they may have a very capable robot worthy of a top 8 placement. This is the only thing I feel FIRST can do for this game.

I agree greatly with this. During the BAE Regional we faced one of our opponents four times throughout qualifiers, and another three times. Truly, something needs to be done about this scheduling.

[JoeXIII'007]

Seeing what everyone is saying about this year's game, I really cannot wait for GLR, Buckeye, and ESPECIALLY the championships... no... really... I can't wait. I just hope there is some more good scoring than I have heard/read of.



-Joe

PS: That said, I am anticipating a more dynamic game as the season progresses.

[Not2B]

I just want to say thanks to all the teams that "went first" this weekend. I'm sure next week will be different, because all of us other teams are now adjusting our thought patterns, and stratagies.

Things will adjust as the season goes on... don't sweat it.

[Jack Jones]

I’m convinced that this game calls for a defensive ramp bot that scores tubes.

If you can’t lift your alliance, then maybe none of you can. You run the risk of losing out on the bonus points, and thereby loosing qualifying matches.

On the other hand, you can expect to see semis and finals where both sides can, and will, get the 60 points. So you’d better hang tubes.

However, no matter how well you score, there are always teams that’ll be just as good or better. No matter how well you score, a strong defensive robot can shut you down if you can’t out maneuver and/or out push and/or out run them. Besides, once you’ve got the opponents outscored, it’s unwise to give them a chance to catch up in shootout. So, the drivetrain is every bit as critical as the ramp and the scoring mechanism.

No offence to the week one teams, we saw a lot of good robots; but you haven’t seen as many in one place as you will at GLR. I predict that many of this game’s detractors will form a whole new opinion once they see that show. I’m pretty sure that the idea that defense, and ramping, and scoring are mutually exclusive will go right out the window.