After week 1 there have been many upsets of the first seeded alliances. This is largely because of the serpentine draft where that last pick is going to be a team that is not always functional. There is almost no incentive in being a top seed at an event because in eliminations you have to have 3 robots that can hang, and by the last pick those robots are hard to find.
2 Questions now:
Will there be teams trying to avoid the first seed at events now?
Should FRC get rid of the Serpentine draft?
No. And, no.
There will be cheesecake… there will always be cheesecake.
Relevant historical context:
2006: https://www.chiefdelphi.com/forums/showthread.php?t=41298
2008: https://www.chiefdelphi.com/forums/showthread.php?t=66374
2012: https://www.chiefdelphi.com/forums/showthread.php?t=105708
2014: https://www.chiefdelphi.com/forums/showthread.php?t=128402
I’m not linking to these to shut down conversation, but rather to give some perspective. I think this is one of those topics that is healthy to revisit every once in a while.
I’ll be withholding my opinion at least until I can compare this kind of data for the past few years. I’m not yet convinced that 1 seeds are worse off this year than they were in previous years.
Now that the importance of the climb is well established, teams will definitely bring cheesecake for the next weeks of competition. I can’t see anyone purposely avoiding the 1 seed. However, if you are in that position, you’ve gotta be sure your 3rd pick can climb, regardless of how deep the field is. 50 points is insurmountable at the moment.
Personally, I think this game lends itself well to producing exciting playoff matches with serpentine draft. I find it far less interesting when it’s very obvious 1 will beat 8, 2 will beat 7, etc. This year, matches can come down to the very end of the match, and I have to say I much prefer it. Serpentine is meant to take away just enough of the advantage of seeding high that playoffs are not a complete blowout.
I think that this year there is a lot more depth in terms of functionality versus other years, considering climbing and gears are the two main points and basically all you need to figure out is a ramp or something to catch the gears and some sort of climbing mechanism. Not saying this is easy at all; it isn’t, but it potentially equalizes a lot of the robots as no matter what you do, you can only carry one gear at a time and place it and you have 30 seconds to climb. This is a lot different than years like recycle rush where there was a huge gap between teams that had stackers like 254 and 118 and teams that just did one tote or bin at a time. I know at Northern Lights Regional this weekend, which my team was a part of, there were a lot of teams that could do gears and climbing. My team climbed every single match, could make it to the top of the rope in less than 5 seconds, did between 2 and 5 gear cycles every match except for one where we played defense, and had auto code, and we still didn’t get picked for an alliance simply because there were so many other teams that could do basically what we could. Serpentine might be an issue with some competitions, but it certainly wasn’t for Northern Lights or Lake Superior.
Powerhouse teams will always want to rank high, and any other team should as well, being able to determine your own fate is very important, especially this year.
Also, a climbing cheesecake shouldn’t be too hard.
Cheesecake the dream
After watching a good amount of the Week 1 Events I have these thoughts (and pseudo-predictions) on the distribution of success in alliances:
There will be more 8 Seed alliances winning events, or at least moving deep into eliminations than in recent years (probably since 2014). While this may seem fairly obvious, it’s worthy to point out.
The game will definitely change. As we progress through the weeks of competition, the higher seeds will be given to the teams who can consistently shoot enough balls to get the 40kP. While skilled gear robots will still see success, the shooters will begin to place higher. To quote Stronghold, breaching robots and low goal cyclers were the majority of successful alliances, with only a few exceptions. The game evolves with the robots, and the shooters can only get better from here.
Hanging is still vitally important, and that will not change. As we saw, the majority of matches were decided in the last 15 or so seconds, even in eliminations. As the caliber of competition increases, those 50 points will still remain valuable, and might even become more so with the increase in fuel strategies. Therefore, the most successful alliances will have three robots ready for liftoff at T=0. However, as people go back to building after these competitions, they will recognized the importance of hanging, and implement a mechanism. In fact, by Week 3 I predict that most competitions will have 24 hanging-capable teams in elims.
To put these random thoughts together into something coherent, I believe that the high seeds will begin to become more valuable as the overall caliber of teams increase. However, I will argue that number one seeds are going to have a tougher time than in previous years, due to their immediate matchup against an alliance that will likely have three consistent climbers. Therefore, that first alliance gear-oriented robot to be picked by the best fuel robot will be vital to the success of the high seeded alliances, and it will be responsible for making up the deficit that might be caused by a less consistent climber.
TL;DR As the overall caliber of robots increase, the fuel shooters will place higher, and the skilled gear cyclers allied with them will have to work hard to defeat what will be an 8 seed alliance of three hanging robots.
I completely agree. Taking it away will only create more problems.
In a 60 bot Regional, IMHO, “functional” is not an issue.
In a 40 bot District, maybe in the first few weeks.
IMHO, the first 2 bots should be Gear placers to ensure the 1 gear auto, and 3 rotor match.
The third bot should be able to climb, score some fuel points (to break a tie), and play defense. Gear in autonomus is a plus (but, I would much rather have fuel scored in Auto).
I’m just going to say that this isn’t how it played out in Hub City, at all.
As has been said there will always be cheesecake. But there is also what I’m going to call the dark horse effect. Sometimes a team gets picked late by a high seeded alliance and they are actually a really good team.
This happened to us in Hub City. We were low on everyone’s pick list (likely because we only had one successful climb on Friday). 1477 and 3847 came over after alliance selection fully expecting that they would have to cheesecake a climber and brakes onto our bot. Then they looked at our bot and went “Oh, you guys have an excellent drive train, we don’t need these brakes.” and “Oh, you changed your rope and now you have a consistent climber, excellent, we don’t need this extra climber” and “Oh, you guys have an active gearage, that’s incredible.”
I’m not suggesting this happens very often, but it does happen. It happened to us in Hub City and I expect to see it happen a couple other times this season. My reason being that a lack of climbs in the first day of competition puts a team low on everyone’s list. However if that team is in a similar scenario to us, they could be low on everyone’s list but actually be a pretty good team.
Rope troubles the first day prevented us from climbing. Our climber mechanism itself is incredibly consistent. Changing our rope brought us from 1 of 8 climbs on Friday to 10 of 11 climbs on Saturday.
So I would encourage teams that are picking to not only look at day one climb data, but to know about every team’s climber. Because they could be in the same boat as us and have a great climber, but a bad rope.
Thought of this thread this weekend in San Diego, when the #8 alliance beat #1 in QF, then won the SF to advance to the finals. Was ultimately beaten by #2 alliance in very close final matches. And this was a large regional (66 teams), with a lot of good robots/teams, so the last pick was by no means “scraping the bottom of the barrel”.
I think we may have more close #1-#8 matchups than in prior years. Have there been any other upsets of #1 alliances?
It’s hard to know for sure if a team is trying to avoid death by serpentine. If a top team is for some reason unable to obtain a top seed and the teams holding the top seeds are un-optimal for an alliance (which they often are in this game) then I can guarantee we will see teams intentionally under preform.
I think this topic would be an interesting discussion on FUN. I personally feel that what makes serpentine alliance selection interesting is that it balances itself out
Wait, hold on a second. Removing serpentine would only give the #1 ranked team the best pick and then 9th best pick. If we really think that the #1 ranked team should be rewarded with a favorable alliance selection, why not just have all captains pick twice in a row? That way, #1 gets the two best picks of the entire event. It would be completely feasible, and perhaps normal, to have a playoff alliance consisting of the best three robots at the competition. Wouldn’t that be cool to watch??
/s
For the record, I like the serpentine draft. Without it, I think the #1 alliance would win too often.
Maybe with some fresh voices. This debate has been a thing for a long time now. When I first heard the argument Tyler repeated a couple of posts above (circa 2004), I didn’t understand it.
My current team has been in the #1 spot a few times now, and it has taught us that scouting is harder than it looks, and that a good pick list must be thought out very deep. Good 2nd picks are much more difficult to make than 1st picks.
However, I still think the serpentine draft is the best solution. Quarterfinals without it would be extremely dull.
On a closely related note, one advantage that higher seed teams often have is better, more strategically valuable scouting data. Frequently, lower seed alliances pick 3rd partners for the wrong reasons, leaving good teams available as late picks. That really helped my team’s alliance this past weekend at St. Joe (FiM). 5567 was really too good to be left as the last pick, but it sure worked out well for our alliance.
You believe the best robots at events shouldn’t win…?