It’s nice to have a rough estimate of how good the other captains are, but I wouldn’t fret too much over exactly where they should go, just roughly.
Its useful for knowing “when to say no”
Example, the rankings are:
1 - A
2 - B
3 - me
4 - C
5 - D
My list is: A, C, D, E, B,
If A picks me - great! Say yes! Lets say A picked E instead.
If B picked me… well if I say no, worst case they take C*, I can still get D which is stronger than pairing up with B.
If my list were: A,C,B,D
If B picked me and I said no - they’ll take C* and I’ll get D - who I decided was weaker than B. So I should say yes.
*assuming C says yes, a scenario specific thing
Ranking captains is valid - especially as not all captains are created equal.
(consider this a teaser of some of the info in part 2)
In addition, you should even look at how they have performed in the individual matches. Not just as an aggregate. You can have 3 teams who have identical numbers when you look at averages or regression. But if you look at their individual data for every match, you might find that oOne team has been getting measurably better every time they play (trending upwards), one is getting worse as their bot wears out (trending down), and the last is widely inconsistent and has some amazing matches and some bad matches.
Consistency in the playoffs is crucial, so that might lead you to going with the robot that’s improving: they might end up playing better in the playoffs than you expected.
OR you might go talk to the teams and find out that the wildly inconsistent team is much better than you thought, and that they’ve been battling a problem that is now fixed. I.e. - a diamond in the rough.
On this topic of ranking everyone regardless of position, we don’t necessarily do that—ex. If we know who our top choices are and we know we’re going to sit at number 1, we’ll decide on our #1 (with backups in case they decline, with at least one outside the top 10), then nix out every other team in the top 8 because there is no point making a decision on those teams because you’ll never play with them. This shortens the overall picklist meeting time to get people to bed sooner. Obviously if things are unclear, we will rank at least through 24/32.
People can argue it’s good experience and such, but we attend 11 events/year where the students can have another experience picking, and I’d rather the students get rest than make decisions that don’t really matter
I’d certainly recommend going at least 25 teams deep on your list. Each of the past two years, we’ve been at events where a top 24 teams withdrew during alliance selection. Additionally, if you only have 23 teams on your list and your alliance partner is absolutely dead set against that last team, you can be stuck.
When we are a top ranked team, I think including the captains above us on the picklist is really useful for the reasons Katie state regarding accepting/declining.
However, at 2018 champs we were ranked 25th and knew we were staying there and accepting whoever picked us. We saved a lot of time by leaving the top 8 ranked teams off our picklist and only creating a 23-team long picklist. For the top 8 teams, we had brief discussions about their strengths and weaknesses so we knew what type of third robot would be ideal for each one.
We usually avoid calls with other teams if we can help it. However, if we have a question about a team, and one of us has contact information for them, a lot of the time we will try to get that question answered. We try to use texts/DMs to get those answers, though sometimes those answers can be complex and a call is easiest.
Usually, those questions are “why did team X break in Y match?”. Knowing that night saves us the trouble of asking them in the morning.
At champs this year, one of those questions was if 1676 was going to pick us. We asked 1676 if they were going to, hoping we collaborate on a list or at least factor that information into our discussion, but they wouldn’t tell us. They opted out of one of the few advantages they had as an overranked #1 seed by limiting our collaboration time to Saturday morning. While I don’t think we would have had better picks with more time working together, we may have been able to come up with better strategies for how the two of us could maximize our scoring ability under defense. (Disclaimer: other than this detail being a bit frustrating, they were great to work with)
I’m sure someone else said something about this in the thread but im struggling to find it.
This may be better for part 2, but when you’re in a basically “guaranteed” top 8 position, do you feel like its a good idea to go to your top team on your picklist and tell them “I want to pick you?” When should you do this, and what cutoffs would you consider it to become a “bad idea” to do so? (maybe past 3rd seed you’d choose not to, wait till the last few matches of the day, etc)
I only do this if im #1 as anything else is fortune telling on who is going to be there for you.
I will ask teams who they plan to pick if they are ahead of me. Usually will get an answer from teams im friendly with, but its never usually early enough to give me any advantage.
This year at NL I tried to get 2052 to divulge who they were picking but did not get any information. However I have a great relationship with people on 2481 so when they decided to go with 1986 as their pick, they told me, and we were able to partner 111 and strategize about an hour before alliance selections with them as they were our 1st choice to pick if 2481 wasn’t going to pick us.
While I understand your discomfort with the idea that a team would change their game to try to impress another team who might pick them, this can be a very valuable way to inform that other team’s picklist. Both alliance captains and those being picked can benefit from understanding if the team being picked has the capabilities to compliment the captain. Sometimes that means playing a particular role. Sometimes that means demonstrating something that you can do but that have not been doing during quals.
Taking 2018 as an example, many of the top ranked teams would run a switch auto during quals to help insure the RP. But in elims, they would run a scale auto. If the alliance captain has a lot of teams to pick from that have good switch autos, they may want to find someone with a solid scale auto that can compliment their own scale auto. So they may ask a few of the potential picks to demonstrate their scale autos so that they can get the data they need.
Now, keep in mind, the prospective picks don’t have to do what the prospective captain is asking. If you chose to play your own strategy for your final matches, that is fine. There may be valid reasons to do this. But also keep in mind that not only are you not demonstrating those capabilities to the captain, but also demonstrating that you may not be easy to work with if you were to be on the alliance with the captain. In either case, whether you chose to do what the alliance captain is suggesting or not, I would suggest that you talk to the captain’s scouts and explain your position. If you don’t want to change your strategy because you think that what they are asking are not your strengths, would cause you to lose those matches or otherwise hurt your team, explain that to them and if your reasoning is sound, that goes a long way to helping those scouts to understand how your team strategizes matches and what your teams strengths are.
One final point (and this is subtle). It is clear from the the post you quoted, that it was 195 that asked 2791 what they could do in their final matches and it was not 2791 asking 195 to change their game. This is part of the game. If your goal is to compete in elims and try to win the event (and this is not necessarily every team’s goal), then the purpose of quals is to position yourself to be a part of the best alliance you can for the elimination round. For some teams this means ranking as high as possible to be an alliance captain. For other teams, it means impressing the scouts from a potential captain and improve their chances of getting picked by that captain. If, on the other hand, your goal is to simply win as many qualification matches as possible (and this is a worthy goal for many teams), then you can simply and respectfully decline any requests from other teams to change your game strategy.
We do everything almost exactly the same. You have to be super quiet about what you’re doing unless you’re #1, then people appreciate you being super open (as long as it’s not a scorch scenario—then I would be very quiet if I was #1).
This is where how friendly/social your team is with other teams comes into play, where more connected teams will get more information and be better prepared overall.
Yes. We place all the teams (including ourselves) on the list, usually captains in their order 1-8 and the rest in the order we would pick them. This oftens leads to discussions of how the initial picking will go, and what alliances we are likely to be up against and how we would need to play to potentially beat them.
More often than not the picks turn out the way we expect. So we’ve already had some strategy discussion about strengths and weaknesses of the alliances we’ll be playing against.
One trick I’ve been using in the past few seasons for Picklists at smaller events is to populate our DNP list first.
If we are at a 32 team event, finding the 8 teams we don’t think are good fits for our elimination strategy for various reasons quickly shows us the 24 teams we have to work with to create a strategy that works.
Usually we do this at the end of Day 1 giving us the second morning to sort our top 24. If we find a team in our 24 is questionable then we look again at our DNP group to focus on who’s the right fit. If a team on the DNP stands out we bring them back into the 24 and remove someone else.
This strategy doesn’t work for every event (regionals, DCMPs, and Champs are large fields), but I’ve found it identifies our 24 faster. If we are seeded higher or have a chance of getting picked in the top four this helps keep us focused on the teams who are more likely available on the back of the draft barring some team sleeps through the draft.
I also like it because usually our scouts identify more DNPs than we ask for so we become more critical of who is on that list and scouts have to argue their reasoning.
Swami_dm - if you are going to follow up on this during section 3 I would suggest using this hypothetical:
- Team B asks Team A what they can do to move up on Team A’s picklist
- Team A tells Team B to show they can play defense against a climb
- Team B does that in their next match and successfully prevents a climb by their opponents
- Team C from the alliance with Team B files a NMIR claiming Team A asked Team B to not return to their hab and it cost the alliance a rank point
On the note of using historical performance to impact picklist decisions-
Greater context of a team (historical performance, you watched them build the robot in the pit Thurs morning, you practiced at their field the prior weekend) is a tool to interpret and read between the lines of the scouting data in front of you but by no means is a replacement for that data. Sometimes you know context beyond just raw numbers that helps to interpret the data.
I realized I never actually responded to this, and I would like to share my perspective on what I consider to be generally very positive discussions. The way I see it, these conversations represent a mutually beneficial partnership. The picking team would like to make the best pick possible and additional information may make this easier or reveal new opportunities. The team hoping to be picked would like to be picked and providing that additional information may increase their chances of getting picked or getting picked by a team they would like to play with.
We provided this information to 195 in good faith so that they could more easily prove to us they were the right pick. They aren’t mind-readers, and aren’t going to know what we want if we don’t tell them. We didn’t want them to change how they played so they would lose matches and would have understood if they didn’t follow our advice in order to win matches.
We wanted the best pick. They wanted to be the best pick. We collaborated to make it happen and it led us to a blue banner together. There was nothing shady here. We also had similar conversations with other teams who we were considering picking about their respective weaknesses and what they could have done to potentially go up on our picklist. We are happy to help teams help themselves and greatly appreciate when other captains are willing to do the same for us.
Question for part 2. What to do about strong opposing opinions between captain and 1st pick about future selections.
I put together a landing page to make it easy to find all 3 parts of the Gettin’ Picky series (and a couple other related resources), feel free to bookmark/share: https://linktr.ee/gettinpicky