Gettin' Picky, Part 3: Getting Picked

Hello! Brian and @Katie_UPS here!

Alliance selection is hard, and there are not many resources out there about how to navigate it. Given the positive reaction to Katie’s thread on scouting, we decided to create a comprehensive guide for alliance selection and everything that goes into it. If you missed it, we highly recommend checking out Part 1: The Picklist and Part 2: Alliance Selection.

Picking is an art, not a science, so much of this advice is subjective and reflects our experiences. We would love to hear about other philosophies and methods in the comments.

About The Authors (a.k.a. “Why listen to us?”)

Katie Widen, self proclaimed Scouting Guru, has been making picklists since 2011. Her work has also been featured by 1675, 3928, 1296, and 253. She takes pride in her obsession with spreadsheets and seeing 20+ students working together to create the Ultimate Picklist.

Brian Maher currently mentors FRC 333 and was a mentor on FRC 2791 from 2016-2019, helping the team overhaul its scouting and strategy processes. Smart scouting and picking helped them take home their first ever regional win in 2017, followed by three other regional wins, six offseason wins, and getting picked in the first round in three championship divisions. He loves getting students excited about analyzing data and scheming up winning strategies.

Gettin’ Picky, Part 3: Getting Picked has been a long time coming and we are grateful for everyone who has read and provided feedback on parts 1 and 2. Thank you for your patience as we took forever to get part 3 finished and published!

Before the event

  • Watch Week 1, and understand how winning alliances play the game and how you can contribute to one.
    • If you are competing Week 1, watch Week 0 and take it with a grain of salt
  • Practice makes perfect. Dedicate a significant amount of time to driver practice before your first event. Driver skill is a huge component of how well a robot performs, and even a robot that can hardly score may get picked with good driving.
  • Develop a pre-match checklist for doing a thorough systems check. This helps ensure reliability during matches.
  • Build a robust robot. No one wants to pick a robot that is threatening to fall apart at any moment. No one wants to pick a robot that can’t take care of itself. No one wants to fix your robot for you. If this sounds like you, here are some tips:
    • Doing more things does not mean a better robot. 9 times out of 10, I’d rather pick a robot that is 100% at 2 game tasks than 60% on 4.
    • Plan your robot before you make it. Design it in CAD - at least somewhat: measure twice, cut once.
    • Use better parts - don’t reinvent the wheel when you can buy it. Use nylock nuts where possible. Use bearing blocks if you don’t have precision manufacturing. Make sure your chain is tensioned. If it’s structural, don’t zip tie or duck tape it.
    • If you can’t program a turret, don’t build a turret.
    • Understand appropriate gearing and torque v speed.
    • Low resource does not have to mean low quality, you can always work smarter instead of/along with working harder.
    • Did you use a KOP chassis? If not, did you really make something better?
      • Brian’s take: most teams who field a custom drivetrain are wasting time they could be spending on building good manipulators, programming, or practicing driving. Plus, the KOP chassis is an incredibly well-tested and robust solution. Many look down on the KOP chassis, but I actually consider it a major plus in potential second picks compared to a custom drivetrain with unknown engineering quality.
      • Katie’s take: I get that using the KOP can be difficult for interfacing with other components so I don’t knock custom chassis - but there is definitely something to be said for KISS. Custom or not, keeping your drivetrain simple and reliable is key.

Be a team that others want to work with/at the event

There are a lot of non-numbers that go into picking, especially when differentiating between a lot of very similar teams on paper. There is value in being the team remembered positively by the drive team, pit crew, or strategy members. Equally, there is value in not being remembered as a team negatively. You want to be an addition to a potential alliance, not a liability.

  • Have a clean pit

    • Know where your tools are. Keep track of your batteries. If you can’t find your tools then you can’t fix your robot. If you can’t fix your robot, you can’t play in matches. If you can’t play in matches, you can’t win them.
    • Low key subset: How many batteries ya got? Having enough batteries to support yourselves during elims is super helpful, having batteries to help out your alliance partners is helpful, requiring others’ batteries to have a successful elims run is a liability.
  • Be prepared and enjoyable to work with

    • How well does your team communicate with your alliance partners? Are you meeting them for the first time as auton starts? Are you seeking them out long before queuing to plan out the match? Does your team follow the agreed upon strategy? Do you come up with solutions if something is causing that strategy to not work? Have you been the team making sure the strategy is executed? Cross-team communication, both formal and informal, is key.
    • Most teams have experienced being in a strategy meeting and being told, “you’re playing defense.”
      • How you react to that says a lot - most teams find it disappointing to be relegated to that position. Reacting with facts, reasoning, and compromise makes you a much better partner than being rigid and seeing options as black and white.
      • If you’re the team bullying other teams into playing certain roles with no compromise, who wants to work with you?
      • Match strategies don’t need to contain rigid roles and all teams need to learn to be flexible.
    • Be realistic about abilities and weaknesses and communicate those during pre-match huddles. Knowing everyone’s abilities and weaknesses accurately allows the strategy team to pick positions that maximize potential earnings and minimize potential losses. It’s okay to have flaws as long as you are aware and work with/around them.
  • Mind your manners

    • This should be obvious - don’t be rude. You can’t go wrong by being nice. Did you win a match you should have lost? Be happy - you earned it, but make sure to be graceful - no boasting or showboating. Did you lose a match? Shake hands with your teammates and, if easy to do, congratulate the winning teams. Did you bring your Gracious Professionalism?
    • Be courteous to others: don’t clog the pit lanes, don’t spill into other people’s pits, return tools that you borrowed promptly and in the same or better condition than they were borrowed to you, etc.
    • Don’t be annoying - don’t stand around another’s team pit without invite, don’t interrupt a team working frantically on their robot for a button, and don’t give unsolicited advice. Also, read the room: some people are not chatty/enjoying your company at this moment, that’s okay. Try to recognize when that’s happening and move on.

Have a robot others want to work with at the event

  • Move and do something meaningful every match:
    • Don’t die repeatedly. Figure out how to not throw chain, not brown out your radio, not unplug a wire, etc. You are surrounded by other teams who have figured this out - ask for help.
    • Earn the easy points in auto + endgame (drive across the line! park on the thing!)
  • Use pre and post-match checklists. Being the robot that breaks down will not improve your prospects. Many failures could have been prevented with checklists.

If it’s good enough for NASA, then it’s good enough for FRC

  • Do Your Thing™ and do it consistently. Try to focus on doing the tasks you do well. It’s better to take your time with things like game object placement/lining up instead of doing it fast and missing/wasting time.
  • Improve throughout the event - small, incremental improvements can increase your chances of getting picked. Get help to improve your auton, find tricks to assist your robot with lining up, practice between matches. If you play defense, watch how others play D and learn from them. There is rarely any benefit to stagnation.

The Sell

  • If there’s a large fix or improvement that drastically changed your performance, it can help to make captains aware of that change, especially low captains who will need to take some risks to pull off an upset
  • Ask what captains are looking for and prove it on the field.
  • On the other hand, Clueless Captains are real. You might be able to sell your wares to them - but here’s the thing, you could tell them you’re the best thing since sliced bread and walk away but you can do better than that:
    • Provide them with your scouting data, show them the stats. You are a better polka dancer than Team X/Y/Z because while your average polka points are lower, you have shown that you have been getting better over the course of the competition - your polka point potential is sky-high!
    • Alternatively, you could help them make a pick list because you are a pro (you read our guides!) and you brought your GP. It’s a win-win, you get practice making an excellent pick list and they don’t stand on the field for 60 seconds with a blank look on their face.

“Why didn’t I get picked?”

This is a great question and a great exercise in self-reflection and evaluation.

In bullet point form:

  • Your data v their data
    • What does your data say about you? What does their data say about you?
    • Do some self-reflection - where did you do well? Where did you under-perform?
  • Different teams prioritize different things -
    • If your robot strategy prioritized hatches but the money was in cargo, then you’re not going to be a top commodity.
    • If you collected data on favorite color and they collected data on favorite food, you’re going to get different results/picks.
    • Your match strategy may differ wildly from other teams - a good post-mortem in the case of not being picked is looking at the different alliances’ strategies and the roles their different picks played. Try to reverse engineer what qualities the teams were looking for.
    • The best thing you can do to hedge against this is show off everything you’re good at at some point during quals.

Miscellaneous notes:

  • Decide when you want to play to get picked: ie if you were to peak at three game pieces in 2019, you should probably play some defense and show off other abilities.
  • If you’re not sure if you’re gonna get picked, and the #1 seed asks you to do something in your match together, do it (if you can). Barring unethical situations like match throwing.
    • Real Talk: Sometimes the top ranked teams are going to ask you to do something (usually play defense) in a specific match (usually with another highly ranked team) to ensure their picking place. It’s probably in your better interests to follow through on that - but don’t act against your own alliance’s self interest, even if a top end team asks you to.
  • There is always some sort of Gut Feel involved in picking… either intentionally or not, especially when looking at the 3rd/4th bots - it’s not easy to place where your team lands on the Gut Feel radar. You can help yourselves somewhat - don’t be the team who’s Gut Feel is “they were really rude to us in the stands.” Be the team who’s Gut Feel is, “They were super cool to talk to when queueing.”

Further reading:
2018 Championship: Alliances by JVN

46 Likes

ok but when is part 4 coming out

but seriously thanks for the effort you guys put into these

12 Likes

If you’re gonna do The Sell, have something compelling to actually say. Don’t tell a captain you can play defense or you had a great match or anything they can gather solely from data and watching matches. They’re already going to take that in and draw their own conclusions, or ask you about it, etc.

The only time I’ve seen The Sell have some value is when discussing strategic options that aren’t obvious from quals. End game strategies are often where this comes up - maybe your robot is uniquely suited for a double Hab 3 climb in 2019 or a triple balance in 2012, and you didn’t have a match together to try it. Maybe you have an autonomous mode that is complementary but rarely seen. If so, talking to captains and maybe getting to the practice field with them for a few minutes could be some help - but you have to start the conversation early to have a shot with this approach.

16 Likes

This is always a hard question to answer. All too often I hear teams say that people didn’t scout well, or something else to deflect. Of course, sometimes teams don’t scout well! I have made plenty of wrong picks myself. But adding onto the great example Katie provided:

  1. The easiest way to drop on someone’s pick list is to break. Saying “it worked last match so it’s fine” is not enough - and often doing 1 set of full systems checks may not be enough! In 2018, on 340, we did a full systems check immediately following every match (except if something broke terribly, which was rare), and then we did robot maintenance. We didn’t want to wait to find out that something had broken without us realizing it. Then, when we were about to be called to queue, we would do another set of systems checks (after we had wrapped up robot maintenance). We only had 2 breaks all year and both were first event. We learned from those and we had a fantastically low stress pit environment for the rest of the year.

  2. Please: don’t build mecanum. It doesn’t help you. There are exceptions to this but I can count them on one hand, and they know who they are. Sure, it can make it easier to score, but consider your honest chances at being picked to actually score in playoffs.

  3. Make sure you are having fun! We have moved teams down on our list because they came off as completely uninterested in pre-match discussions. This year at Buckeye, we played with a team who was absolutely thrilled to be playing and were so excited to do the strat we proposed. They did a fantastic job on the field and we moved them very far up on our pick list because we had a great time with them and we felt that they would be a great partner to trust with a job in elims. While we didn’t end up picking them, we absolutely remember them and would be happy to play with them in the future (6936, if you’re reading this, keep it up).

8 Likes

The wait was more than worth it!

3 Likes

If you want help getting picked - build relationships with influential mentors on other teams outside of competition settings. Not only will you learn as much as possible from them, but building friendships is a good way for people to know if someone is reliable, trustworthy, capable, and cooperative.

When I was with 11, 3929, or 5686, we always had a better shot of being selected because we were friends with some teams outside of competitions (in addition to having a capable robot). People want to work with their friends, especially if all else is equal between robots and on-field performance.

10 Likes

It’s also just really great meeting new people and learning from them. This year I reached out to @aciarniello before Buckeye to talk about autos, since there was a good chance we would partner up anyway, so I wanted to have a good shot at the both of our teams breaking the curses. In the process, we both learned a lot from each others teams and I made a friend that I enjoy going to visit at championships and IRI.

The FIRST community is great - there are so many great people to meet and so many wonderful stories to hear.

2 Likes

I can’t recall a case where I’ve ever really factored in a team’s sales pitch to us. Seeking out an alliance captain to pitch them seems like a generally unproductive task. Although I remember a tale from FRC antiquity about Raul from 111 pitching a captain on how they’d beat them in the playoffs if they weren’t picked, and then the captain picking 111 instead. But, like I said, that’s a tale from long ago in FRC, and most of us aren’t on Wildstang in their prime.

I do think having info sheets about your robot to give to scouts and alliance captains can be helpful. It can provide a quick shorthand to help them remember who you are, and what sets your robot apart. It may also impart information that they didn’t think to capture in their pit scouting, or information on lesser used niche features you may not have been able to feature prominently in qualifications (such as configurable autonomous modes or collaborative end-game friendly design features).

I am interested in the reverse scenario, though. There have been instances where scouts from high ranked teams have approached our team for additional scouting and follow-on pitches, especially at the Championship event. I know in one scenario, this pitch was instrumental in our selection (although this was 2007, when scouting was far more rudimentary and less thorough than it is today). A team that had been impressed with a recent match and had our scouting sheet, but wasn’t sure how we would play in more physical, defense-heavy matches. They asked us about this, specifically. Fortunately for us, we had received a “toughest robot award” from another team in our division just a couple hours prior to that, and showed them that trophy to help demonstrate that we were ready to play through hard defense. I can also think of a couple other times in more recent Championships when high ranking teams can to do very thorough follow-on scouting of our robot in the pit (but both of these teams were ultimately selected onto the #1 alliance while we ended up as lower ranked alliance captains).

Minor gripe. The roboRio browns out. The radio resets. Stage 1 roboRio brownouts out are a momentary thing that will cause jitters or stutters when driving, but should not ultimately cripple a team (unless there’s an issue somewhere in the power system that causes them to be sustained over a longer period). Radio resets will leave a team motionless on the field for 30+ seconds as the radio reboots and establishes comms with FMS again.

1 Like

If a high-ranked team shows up in your pit, they become priority #1 for one or two people, and priority #2 (behind robot repair and/or match strategy) for everyone else. Failure to do that can cost you in playoffs. They’re there because they are interested in picking you.

1 Like

The Sell has worked on us before, though rarely it works outside of first pick and the team already has some relationship established. Building a good robot with a good driver and the right features is the easiest way to get picked, maintaining good relationships is an easy #2

Seconded. I’m hopeful for the day that somebody puts together data across multiple games and all regions that show the correlation between choosing mecanum drive and getting picked later in the draft (or not picked at all).

^ This. On my old team we ended up winning an event thanks to a double HAB 3 climb that was not obvious at first. Talking with the team who ended up picking us before hand made a really big difference, so don’t sell the sell short. If you have something important to bring up that could help your potential alliance, go for it.

That was my bad. To be fair, I was referring to when the radio connection is loose (and don’t have better words for it) and not actual battery power but both are relevant, common failures that teams can avoid.

2 Likes

I’ve had a slightly different experience when it comes to “The Sell”…

I sell 4607 all the time! Some years I’m selling a rusted out 1972 Ford Pinto and other years I’m selling a nice new Porche… the process is always the same and almost always effective.

It’s important to be honest with yourself and other teams. Accurately assessing your own abilities will enable you to accentuate your best features. You should know what you’re good at going into an event, and determine match strategies that feature your team doing what you do best. More often than not this is also the best strategy for your alliance’s chances to win.

In both 2018 and early in 2019 4607 prioritized strategies that maximized game piece throughput. The idea being that most teams primary separating criteria for teams will be the number of game pieces they score in a match. It might seem like an obvious strategy “score a lot” but there were definite tradeoffs including not going for bonus Ranking Points in some matches where they might’ve been possible.

When pitching 4607 to other teams I always talk about upcoming matches and what we will do, not what we’ve done. Anybody I want to be picked by already knows what we have done. It’s important to be aware of your team’s potential role on a playoff alliance and sell to that role. In 2018 4607 was an ideal 3rd robot on playoff alliances. In 2019 we were a good 2nd option. Understanding what each role means and what you need to show in quals to fit each role is important.

One other critically important note about “The Sell”… If you’re beginning your pitch after your last quals match you’re already too late. I’ve started “pitching” teams before the build season even begins. What this pitching amounts to in the end is networking and relationship building. Having a connection with other teams will naturally increase their level of trust in you, and thus make them more likely to pick you.

I believe “The Sell” is a valuable part of the FRC strategy experience and can lead to positive results if done the right way.

3 Likes

In my experience, “The Sell” is better when you take an approach closer to a market survey. Early on in the event(or even season), figuring out what top teams are looking for and trying to shape your event game plan to fit what your customer wants. Best case you’re successful and find yourself on a great alliance, worst case, you’ve networked with teams and figured out how you can improve and execute for the next event.

Selling a nonfunctional or undercapable robot to an alliance captain is exceedingly difficult, but figuring out how to walk away from the event knowing more than you did before can change your team for the better.

Be confident in your current capabilities but also realistic about your shortcomings. I respect teams who are upfront about anything that might affect our matches together.

4 Likes

This is something we really pay attention to when picking. It is also important that teams have enough chargers to keep all their batteries fresh. Elims will hopefully run long and we want partners who are prepped for the quick turnarounds.

3 Likes

I agree, how many batteries a team can charge at one time is one of the few pit scouting metrics that matter to me.

1 Like

Sometimes its important to think about the level of competition at your event and if that will shift the typical alliance meta. This is something that I (1710 alum) missed when we were at IRI in 2018. Knowing that we were not even close to being a top scale robot at that event from the pre scouting we did on the teams there (IRI Pre-Scouting), we totally switched our usual strategy to only focus on switch, exchange, and defense to try and stand out as a solid third bot.

What we failed to see was that alliances at IRI stopped really caring about those tasks as the scale was king. It wasn’t that we didn’t execute well either. I don’t have the data handy but I remember we were one of the top teams on total cubes scored and had several strong defensive matches, but it didn’t matter because we didn’t adjust to match the shifted meta. If we had asked some of the top teams there they might have indicated that what they were actually looking for in a third robot and matched our strategy to that.

This is something that happens a lot as the level of play increases from regional and districts to championships so its important for teams to determine if they need to change their strategy for each event.

4 Likes

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.