2018 Championship: Alliances

With Championship Divisions being announced soon, robot strategists everywhere are thinking through the same hypothetical question: "Who do I want to work with?"  The folks here at the House of Black are no different...

The most common question we get is: "Dude... are you gonna rip our bumpers off?"

But, the fact is, the second most common question we're asked is: some version of:

"What does a team like 148 want in a partner?  Who do the Robowranglers want to play with?"

I'm happy to spend a little time answering this.  Some of it may be obvious, but I know not all my readers are master FRC strategists.  So, here we go...

I should start by saying - I'm a terrible scout.  A proven, bad scout.  I don't make the pick lists.  Truthfully: I am barely asked for input on the pick lists.  I'm a strategist, and a coach.  So I can talk about the strategic priorities, and explain things from that perspective.

That said, I'll do my best...

IMG_8564.JPG

What we want in a partner changes a bit based on the situation.  There are a few different ways things play out, a few different cases to discuss.

"We'd love to play with y'all!"

The first way this plays out, is that we get picked by someone seeded higher than us.  We're pragmatic folks here for business, so of course we need to decide if we have a better option.  If we think we can end up picking someone better, we will respectfully decline.

What constitutes a better option?  What criteria do we judge the team that is picking us?

Simple, we judge them by the same criteria they judge us, and by the same criteria we judge our potential first-round picks.

First Round Selection...

What are we looking for this year?  Ideally, we want someone who's much better than us.  (We want to ride some serious coat-tails.)  For real:

  • We want to be with an "elite" level scale robot.  (That way we can let them do all the work, and we can play some defense.)
  • Hopefully they can place more than 6 cubes on the scale before they start to struggle and drop them.
  • We want multi-cube autons, with options.
  • We want someone that picks up cubes fast, from anywhere.  Portal, Pile, Home Rows, Open Field.
  • We've got our Robot Wrangler, so we don't care as much if they have a "buddy climber"
  • If they do have a buddy climber, that's nice - we'd have some options!  We're pretty "buddy climber compatible" so we'll probably be able to get lifted with them.
  • Is there a spot where we can attach our rope?  How do they feel about getting Robot Wrangled?
  • It'd be great if they had a fast single-climber, ideally one that is kinda compatible with our climber.  We can put out hook up on the bar, get out of the way, and leave room for someone else to climb before we winch up.  This works way better than it should, considering it was not one of our design goals.
  • So, yeah: lots of climbing options.  We lift them.  They lift us.  We each climb on our own...

That list above?  That's also who we want to pick us.

Why should someone pick 148?

Also, that same list above.  I mean, we designed the robot - so of course Uppercut will (attempt to) do the list of things we like to see a robot do...

Maybe you have a different list?  Maybe we fulfill some of your needs?  Call us.  Here's a scan of our card:

148-Business Card.png

Not Good Enough

"Robowranglers, you don't do the things we want you to do." - Fine, pick someone else.
"There's no one better." - Fine, pick us.

Best Bad Idea.gif

As veterans know, it's all about finding the best option since the perfect option is not always available.  We work our way down the priority list.  We talk through trade-offs.

At some point, maybe we realize we need to change our criteria.  For example "we really want a great scale robot" becomes a problem when the best scale robot available can only do 1 cube per match.  Maybe at that point we should stop worrying about scale performance, and restructure our plan around a different type of partner capability?

What about if we run into two very similar robots, one is a little bit better at the scale, and the other is a little bit better in the end-game?

What if we encounter two robots, one is better than the other... but has some inconsistency.  The second is very consistent but lower performing?

These decisions are very situational.  How much better at the scale?  How much inconsistency?  Sometimes strategies require very specific levels of performance.  It depends on what's available, and how deep the division goes.

The details matter -- this is why we scout every match, and take stats on every team.

Blah, blah blah... Really - WHO do you want to work with?

You probably know the list of names as well as we do.
We'll see how division assignments play out, then we'll see how match schedules play out, then we'll see how things play out...

The interesting thing is, it seems like people assume teams like the Robowranglers go into the Championship with a pre-conceived decision about who we want to play with.  That isn't really true.  On-field performance, scouting, and stats dictate 99% of our decision-making.  You think we'd just pick a team based on name recognition?

To quote one of my former managers: "You've done some awesome things, we're not disputing that... but WHAT have you done for me LATELY?"

If there are two robots of "about equal" performance, absolutely the tie-breaker goes to our friends. Most people think that is the first filter, but really it's one of the last...

BF.gif

At the end of the day:
We love playing with old friends.
We love playing with new friends.
Do you want to be our friend?

Second & Third Round Selection...

Here's where the money questions start coming up.  This is the topic which really inspired this blog...

"What do you want to see in a 2nd robot?"
"Yo, JVN - what do you want to see from us today?"
"We know your scouts are watching... what are they looking for?  Anything we can show them?"

Good questions.  A lot of this depends on how deep the division goes.  Divisions at the Houston 2018 championship will be 60-70 teams.  How much firepower will we be able to get with a 3rd or 4th round robot?

Let's go through the list of what we're talking about when making the decision...

First off - when I say "They can..." what I mean is: "They have demonstrated capability such that we'd ask them to..."
Big difference.  Lots of teams can score in the scale (when the moon is right, on the fourth Friday of every month), but only some teams have shown performance on the field that makes our (very snobby) scouts say: "Hmmm, yes... I think we'd ask them to help with the scale."  Keep that in mind when you see "someone that can..." on the list below:

  1. Consistency:
    Drive reliably, every match. Never "die" at any time.
    No thrown chains, no dropped bumpers, no lost comms, no popped breakers, no low batteries, no broken wheels, no excuses... NO EXCUSES.

    (Did #1 sound familiar? I stole it right off my Priority List from week 1 of the build season!)
     
  2. General Performance:
    -Can we get someone capable of "mixing it up" with opponent robots?  A little defense?  A little "hold your ground"?
    -Can we get someone that we trust to play under defense?
    -Can we get someone who can play on the far side of the field, without getting "lost" or getting penalties?
    -Can we get someone who is willing to work with us?
    -Can we get someone who will "go the distance" in the elims?
    ---Do they have enough batteries for  long day of match-play?
    -Do they "play smart"?  Does it seem like they can fulfill more than one role in a single match?
    -Do they have moments of: "What in tarnation are they doing down there?"
    -Are they easy to work with?  Do we have information on what it's like to work with them?
     
  3. Scoring / Teleop: How deep will the division go?
    -Can we get a "good" defender?
    ---Do they have the ability to hold their ground?
    ---Does it seem like they're "in control" when moving around the field?
    ---Have we seen their driver demonstrate "good instincts" about how to defend a team?
    -Can we get someone that can hold a cube?
    -Can we get someone that will put a few cubes in the exchange?
    -Can we get someone that can score significantly in the exchange?
    -Can we get someone that can score in the near switch?
    -Can they manage both the near switch, and the exchange?
    -Can we get someone that can score in the far switch?  Are they good at "portal to switch" cycles?
    -Can we get someone that can score in the scale?
     
  4. Auton: How deep will the division go?
    -Can we get someone that will cross the line every time?
    -Can we get someone that can line up on 1 switch and score 1 cube in it if it happens to be the right color?
    -Can we get someone that can start center and score 1 cube in the switch?
    -Can we get someone that can score 1 cube in the switch and pickup a 2nd cube?
    -Can they score 2 cubes in the switch?  (Etc, etc)
    -Can they score a cube in the scale?  (Surely, the division won't go that deep?!?)
     
  5. End-Game:
    -How Robot Wrangler compatible are they?
    ---How did they respond to our folks asking "How do you feel about riding the Robot Wrangler?"
    ---Are they a microwave, or a fridge?
    ---Do the Robot Wrangler forks fit under their chassis?
    ---How heavy are they?
    ---How tall are they?
    ---Is there a good spot to attach the velcro rope?
    -Are they good at driving up on the platform?
    How good?  Like... "they don't even notice the platform" good?  Or more like "with a nudge, they can hypothetically get up on the platform, but we haven't seen them do it" good?
    -Have we seen them do "buzzer beater" plays?  (Have we seen them score a last second cube, then run to the platform while avoiding a defender and quickly hop onto 1619's lifter?  That indicates they might be good at buzzer beaters.)
    -Do they have a climber?
    -Do they have a buddy climber (that we'd actually use)?

Priorities / Tradeoffs

As I said before, everything is tradeoffs.  At some point even if scoring on the scale is "more important" to us than scoring in the switch, a team's proficiency at scoring in the scale might get low enough that we'd prefer a robot which is better at the switch.

Scale Robots vs. Switch Robots

"JVN said it!  He said it on the blog!  The Robowranglers want switch robots.  They say scale robots aren't as good on the switch!"

How.  Ridiculous.
I don't understand why people espouse stuff like that.  We judge robots based on what they do, not based on what they design to do.  Just because a robot has an elevator of some kind, we don't subtract 10 points off their switch scoring capability.  There are plenty of great switch robots that can also score in the scale.

We might pick a robot which scores in the scale, but then ask them to ONLY score in the switch.
"Look, you guys are MONEY in the switch... your scale game is just distracting you.  Let's set you up so you can do what you do best." - Happens all the time.  Works out, every time.  Build strategies around actual robot capabilities, not wishful thinking.

"JVN, who is the BEST switch robot... is it 7179?"

Beatty Krunch is amazing.  I'd love to work with them.  They're definitely among the best in the world.  However, if I needed to choose a "best" switch robot I'd probably go with 254.  I'm really hoping to get them as our 3rd robot.

It's an unconventional choice, but I think they'll still be around.  After-all, most people believe scale robots are bad at the switch.  That's why the Poofs are my sleeper switch robot of the season.

3rd + 4th Robots + Cheesecake?

At Championship, the fact that each alliance has four robots opens up some interesting possibilities.  If you're on the #1 alliance, you pick your 3rd and 4th robots back to back.  How does this effect picking strategy?  Sometimes, it doesn't at all.

Sometimes, it allows you to pick two robots which specialize in different roles.  "Team A is better at defense, Team B is better at exchange scoring." - "Whatever, they're both great.  Let's pick em both and let the strategists figure it out!"

In some years, this has also opened up options for "robot upgrades" commonly referred to as Cheesecake.  Teams will pick a 3rd robot to play, and pick a 4th robot who they know is willing to receive some upgrades.

Sometimes these upgrades are functionality additions (i.e. adding on a "Can Grabber" to a team in 2015) but more often these upgrades are just robustness improvements.  "We picked you because you guys are AMAZING, but you die halfway through some matches.  before we put you on the field, we'd like to try to make sure that never happens again... if we work together, we can chase down this gremlin and get you up to the 'every match, no excuses' performance level.  You guys feeling it?"  Heck yes, they're feeling it.  (Note: we ask ahead of time if they're willing to "feel it" - we would never put a team in this situation without first confirming they want help.)

With these upgrades, the 4th robot is more robust and ready for later (tougher?) matches.

For Championship (other than the velcro ropes for the Robot Wrangler), do we have any special cheesecake prepared?  No, not really.  We've got some small things, but we don't expect to need them.

The Talk

Okay... things are about to get real.  Brace yourself for some hard truth.  This is a 148 perspective, but I know for a fact this is not just a 148 opinion.

"Hi, I'm John from team ABCD!  We noticed you're seeded #1 right now!  We've been watching matches, and we think we'd be a great partner for you guys!  Where are your strategists, can we talk to them about working together?  We think you should pick us because..."

Does this ever work?  No, it really doesn't.  Save yourself the effort.

Somewhere along the line, people got it in their heads that "if we want to get picked, we need to sell ourselves."  I've been on this side of things.  I've tried to sell a non-performing robot to a top seed.  It.  Does.  Not.  Work.

Sell your robot, on the field.  The scouts from 148 (and basically ALL teams) are watching every match.  If you want to sell your capabilities, show it out there.

NOW... if you have a specific message for the scouts, by all means bring it to the pit and they can pass it along.  "We had a bad main breaker in match 22 and 45.  We fixed it.  Our performance got a lot better after that.  Please let your scouts know what happened." - This is FANTASTIC information, and we will certainly take it into account!  No other editorial needed...

SIMILARLY... if you want to know what we're looking for specifically (beyond the above list) you can come by and ask.  Though, honestly... our scouts are usually on top of this enough.  We very frequently send people by to ask questions about capabilities if there is something we want to know/see.

Answer the questions from our pit scouts honestly.  Show your capabilities on the field.  Let us know if something weird happens that you want us to understand / consider.  This is what we do, this is what we wish everyone did.

If you come to sell yourself, you'll probably hear something like this:
"Yeah, you guys are great!  I'm not part of the decision making process but I can tell you our scouts watch everyone.  I don't know exactly where you are on the list, and I don't know exactly what type of robot we'll end up with.  Anything you want me to pass along to the scouts?  Oh, you fired your driver because he got into trouble?  That's... interesting news...  We'll be watching your new driver very closely."

PS - The "we watch everyone" sounds like BS, but I promise it's not.  Yes, this is a blowoff speech - but at least it is 100% honest.  The scouts are busy.  We don't want you taking their time with a sales pitch.

"How do I get a 148 shirt?"

The students have some of our grey trading shirts.  I really don't know if there is a "best process" to getting ahold of one of them.  I wish there was a more efficient way, but the best advice I can give is: "Find a kid that looks about your size, and ask if they want to trade."  Don't be offended if they don't want the trade, economics can be brutal...

"JVN, can you get me a black Robowrangler shirt?"

Honestly - no, not really.  We set a rule several years ago: "We do NOT trade the black shirts" and to try to keep things fair I try to abide by that rule.  There are only a handful of extenuating circumstances where I'll violate the rule, and even then I need to ask "the boss" for her permission.  That's why, she's the boss...

EDIT: We built a site where people from the community can submit their interest in trading for grey 148 shirts.  Check this out.

Who do you expect to work with this year?

Honestly, no idea.  Like I said... we'll see how it plays out!  Lots of things will leap into motion when divisions are announced, but nothing I wrote above will change.  I'm still hoping to get 254 as our 3rd robot.