If you can find a recording from SVR, you’ll see what we’re talking about. Teams picking weak teams for their first pick, trying to opt out of captainship, and the list goes on…
Lower ranked captains were in a good position to create strong alliances, but failed to do so. Good teams are rarely left for the 1st and 2nd alliances, however at SVR, good teams, were widely available.
One reason, people focused on minibots. Well at SVR it was autonomous that won the finals not minibots.
I so wish I could have been there, MTR would have crushed alliance selections; the competitions we have been at have been pretty good for selections. When I said nothing is wrong I probably exaggerated a bit, because I have noticed a few lower teams getting picked at regionals/districts , but they were raked low on our data. As far as how high those teams ranked according to FIRST I have no idea what it was.
Also does anyone have video from SVR? I would like to take a look at it.
A big problem with FIRST’s ranking is it favors consistent minibots. If you get first place in most Q matches you will win most Q matches. However come the finals, its a very different story, and sometimes very hard to predict. At SVR the field was starved for tubes, so ubertubes accounted for huge portion of the tube score. 254’s alliance did not have the fastest minibots, but put up 2 or 3 ubertubes vs 1868 who had the best minibot, but their alliance only put up 1 ubertube. The match was over before the drivers touched the controls.
The game that ends up being played on Einstein is going to be crazy interesting.
I cannot comment for FRC but I think alliance selection can get disappointing in FTC. Despite different competitions I do see some trends in both competitions in this subject of matter.
In the NorCal FTC tournament, a few outstanding teams did not get chosen for alliance selection and a few bad teams were chosen instead. Part of this is due to the randomness of competition. Even for teams considered the best, things don’t always go as expected.
I think the tube starvation situation is one that contributes to the randomness, so it requires high levels of strategy.
While I’ll agree that minibots were overemphasized during the alliance selection, I would not say that it was ubertubes that won the finals for 254’s alliance. If I remember correctly, during that last match 1323’s alliance was 1 tube away from completing a logo on top, and one team had dropped or failed to place that very tube several times. I would argue that being able to place tubes reliably in general, uber tube or normal, was what decided that match.
254, 1868, and 1538 all used the same minibots.
I appreciate the all the nice things said about my team’s performance at SVR. I am glad we weren’t the only team who thought we were good. We were surprised we didn’t get chosen, sort of.
We knew our minibot wasn't the best, it was a pathetic 3.5 second climb, but we did successfully deploy it several times during qualifications, winning us at least one round. However the refs were very inconsistent with how they judged us so they sometimes counted our mini-bot but they disabled our tower twice, both times they were wrong and we were able to cite the rules and prove how it could physically not have been above the line, costing us one or two wins.
We could have also won another of our matches if we had completed our second logo (we had the triangle in possession) rather than going for minibot, which failed do to a hardware failure, basically our minibot got stuck on our hostbot.
We had actually been working on our autonomous using dead reckoning and line sensors. But were slowed down when we discovered that our encoders on our wheels were different counts per rotation, which explained our turning on the practice field while testing. We also had been rather hesitant after our first practice round at LA where a bug in the code caused us to start smoking during autonomous as we roasted a motor.
Our team was sort of depressed when we weren't chosen, as this is our last regional and this was a large portion of our team, including myself, are graduating. But we can always do better during the offseason competitions and next year when we will be using our swerve drive ;) (which is made just we didn't have time to implement into the robot), that way we won't be able to be defended anymore. But overall we are proud of our robot, we had good driving, a decent minibot, and a good scoring robot. We learned a lot this year, and we will improve our performance next year. In the end we did our best, learned a lot, and had fun. With that said winning really doesn't matter too much to us.
We were surprised by the lack of scouting/preparation of the alliance captains at SVR. Even though we knew after friday that we would not be an alliance captain, we still had a team of 6 students in the stands scouting for us, when we brought only 12-15 to the tournament. My team believes that the lack of adequate scouting and preparation of alliance captains is why so many good teams were not chosen for eliminations. I noticed that a some of the lower seeded alliances picked the next person down on the list rather than a team that would have been better for their alliance.
Again thanks for the compliments about our robot’s performance.
I have so far been to five events this year and one thing that is consistent are teams in position to pick who are obviously unprepared to make a pick and are just overwhelmed by the moment and a shallow pool of teams to pick from.
This year’s game is very challenging with all of the tasks that have to be done and the potential for crushing penalties it magnifies the potential for stupidity and there are many teams eagerly embracing it.
At every regional I have been at there were several teams in the elims who had no business playing on Saturday afternoon. Next week we will see MUCH better quality of matches in Michigan and Philadelphia and see this games true potential of how things will be in St. Louis (for the most part).
http://www.chiefdelphi.com/media/papers/2370
Here is a thread I started a year or so ago about scouting methods. One thing it does not cover in great details is “The Pick List”. I will post up a template that we use. Every team Friday should make a list of at the very minimum 1-23 (24 if you include yourself). Teams that you would be interested in having as a partner. With detailed scouting, this usually still requires an hour of debate mostly trying to flesh out the slots for 12-24. This is the most difficult area as this area typically has the smallest differentiators between teams.
On our team, we also include a bubble section of around 4 teams, and a DNP section (Do Not Pick). DNPs usually result from gathering an abundance of general penalties, DQ potential penalties, and/or often not functioning or broken. Here in Michigan, this list is usually quite long at the first event, and much smaller at the second as teams get their controls sorted out, and most teams learn the rule better.
Saturday morning, we do a walk through on the bubble teams to see how they are doing, and discuss any issues we may have noticed on Friday. If there are specific behaviours we would like to see from a team (tube herding or defense), we will talk to them about their schedule and ask them to exhibit this behaviour in a match of their choosing. Some will agree to this, some will turn us down. We then refine the list right up until the end. Make a duplicate, and send it out. The duplicate is then used to discuss selections with you partner. Typically off to the side.
Even knowing all this things, we are still often caught of guard for the alliance selection process.
Part of the selection issue is that with the WLT system, ranking tend to fluctuate a lot right up until the very end. This makes it very difficult to discuss strategies with potential alliance partners. Also, because team captains are called to the field immediately after the last match, there is very little time to adjust lists relative to rankings.
I would love for teams to be given a 10-15 minute break after the first round to discuss next round picks with their scouting teams and new partners. Yes this would add some time into the alliance selection process, but it would be so beneficial to the overall competition. As the captains out of the field are frequently members of the competition team, they haven’t seen enough matches to make informed arguments with their other partner.
As a thought of discussion:
What if there was the “Wheel of Fortune” picking rule where you loose your turn if you pick a team that is already in an alliance? Would this reduce the ill-prepared, or would it just be adding injury to embarrassment?
Another thought. If you notice that a young team is doing well enough that they could be in the position to be an alliance captain, go up and remind them of the importance of making a pick list. If you are “mentoring” another team, make sure they do this.
That’s because there is no real single way to scout. Many teams do it different ways. Some teams go strictly off the Scoreboard and ranking. Others take other things into account. Some do both. Some teams run a server with a dedicated scouting team in the stands. We’ve done this with great success since 2008.
There isn’t a right or wrong way to scout. It really comes down to what factors are important to your team.
Well the worst I heard was someone told me a team was picking down the list on the screen and was turn down by every team and eventually ended up picking themselves. Right then and there someone probably should have stepped in to help them.
There is a wrong way: Not scouting at all, like the team who just picked down the rankings.
Worst form of scouting ever:
“Anyone want to get picked make some noise!”
This actually happened at Buckeye in 2008.
That actually just happened at SVR, leading to this discussion. We have had some misplaced teams and head-scratching picks at Philly but noting that bad. Most of the questionable picks are made with quickly, so there must have been some thought process behind them.
I was the team representative for team 1868 for alliance selection, and we pushed very strongly to select 691 since we had worked well with them in our qualification matches. We had covered this choice with 1323 beforehand and agreed they were the best choice.
However, when I was on the field the representative from 1323 bypased our list (which had 691 on the top) and instead went with what one of their mentors was shouting at her from the first row of seats-604.
However, I think this was a minor concern compared to the number of declines in selection. The rankings at SVR were so strange and jumped around so much that there were teams who did not expect to be anywhere near picking position shoved in, and I think this was directly related to the amount of confusion.
One thing that makes alliance selections particularly difficult this year is the need to assemble an alliance which can effectively play every aspect of the game. Most teams, even if they’re unprepared, understand that they’re looking for robots which “do it all”. Typically those teams will either decline a pick from the types of teams who are unprepared to make the selection, or there aren’t any left by the time they get to make their pick.
I believe this is also the reason for the large number of declines. Teams who “shouldn’t” be in the top 8, or should but are unprepared, are picking teams who they perceive as “Good” without giving full consideration to whether the teams complement each other. A team with great tube scoring but no minibot and no ubertube that ends up top 8 will probably be declined by a team who also doesn’t have a minibot, because they know it will put them at a disadvantage.
They picked us
Callie, talking things over with 1323, they told me that they were concerned about the reliability of their own minibot, so they wanted to choose someone with a fast, reliable minibot. 604 fit this bill, 691 did not.
my sympathies.
I have to admit, being on that field saying “Team 2159 graciously declines the position of alliance captain” was probably one of the most hilarious moments of my life. Firstly, I’m not even the team captain of the RoboPirates-- I’m simply the driver that was called to be the representative of our team at the last second simply because our own captain was not feeling too well about the whole moment.
As a small, 6 member team in the finals, we were simply overwhelmed by what the judges said. Did we scout before we were called up to be an alliance captain? Yes, and no. We were solely a strong, defensive team, and we looked for teams that would balance our alliance. We had not been able to finish the other features of our robot, so we were forced to lean on our strongest feature: our two speed transmission. Team 691 and team 3256 would have been my top choices IF I would have remembered their team names. However, our mentor told us to simply decline being an alliance captain so that we would not have to worry about picking teams.
Why did plan on declining the position of alliance captain? Well, a couple of years ago, our team was an alliance captain at the Davis regional. We declined the position then, BUT the judges still allowed us to be picked. In fact, we were picked by the Fembots, team 692. I was stunned when they told us that we would not be able to compete if we declined, so I was forced to turn to my handy-dandy SVR Flier that I picked up at the last second before running on the field.
You’d be surprised at all of the people that came up to me after the alliance decisions and asked what was up. After telling them this whole story, they understood completely. Do I regret picking team 100 and team 766 for our alliance? Nope. They’re all great people who I had a lot of fun with. In fact, I want to thank our good friends at team 100 for getting me out of that mess. Sure, winning is awesome, but to us, FRC is about having fun, making new friends, and spreading the values that FIRST teaches us all-- most importantly gracious professionalism. Let’s keep it that way.