Chief Delphi - Alliance Request

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)

- - Alliance Request (http://www.chiefdelphi.com/forums/showthread.php?t=146561)

Re: Alliance Request

Quote:

Originally Posted by Aura_ (Post 1566300)

Thanks, but when we get constant horrible match ups how are we suppose to advertise ourselves any better?

There are two situations with of "horrible match-ups."

First, your alliance partners are "horrible." If you feel that is the case, find out what your alliance partners can do and fill in the gaps with your robot. For example, 3661 is a chassis bot. At Wake County, we chose them as a defensive robot because of their speed. We won the event with them. At Campbell they stopped us and 2655 from capturing the tower in qualifications. In eliminations, we chose them again as a second pick and in one of their matches they only allowed two low goals the whole match (once they got back). Teams have potential, even if you believe their one of the worst teams at an event; you have to recognize that in every team to be successful.

Second, your opponents are very good. Yes, of course, you will definitely play against multiple top-eight teams in one match in qualifications. Again, recognize how your teammates play and develop a strategy that counters the opposing alliance. Even if your alliance does lose the match, if you perform the best that you can, you will be noticed by other teams.

Onto talking to other teams about forming alliances. Usually, if we're on the top 8 (particularly the top 4), we'll approach the team that we want to pick and tell them that honestly, "we're planning on picking you guys. We think we would together good. What do you guys think?"Usually since we pick partners that work well with us for strategy, they'll usually want a robot like our on their alliance. (As a breaching robot that likes passing to teammates with occasional low goals, we generally want a reliable high goal shooter that we can easily funnel boulders into. Generally, high goal shooters want to shoot as often as possible, so by us demonstrating throughout qualifications our passing and shooting, usually high goal shooters want a robot similar to ours.)
It is quite often we're not the top 8. At Campbell we were 6th, but had discussed strategy with 2nd. We introduced ourselves to them and discussed how we could help benefit them. Due to our qualifications, they had noticed us performing well with boulders. Right before alliance selections, we did end up deciding to not form an alliance because we couldn't get the tower done fast enough.

Point: If you specialize in a role and perform your best in that role, other teams will notice.

Re: Alliance Request

Quote:

Originally Posted by Dezion (Post 1567358)

I would also caution team members to keep your attitude in check even if you feel like you are constantly being matched with "horrible" teams. I'm sure we have all been on both sides of that coin. Learn to see the positive in every scenario, and make the best of the resources you have. Maybe they're a new team and legitimately don't know what they're doing. This is an excellent opportunity to give them pointers on how to play better, how to play good defense, etc. Don't complain about them dragging you down - do your best to pull them up.

If your team is seen to be vocally complaining (or bragging) about carrying the match when paired with useless partners, you may get knocked down the pick list for being arrogant or potentially hard to work with.

Re: Alliance Request

Here is the elimination list by qualification rank in week 0.5 color coded

Notice #49 got selected. Why? (probably because they had a unique quality or showed something)

This is typical what it shows is...

Top 16 went to eliminations

5 more in next 15

3 in bottom third

17 and lower at risk for not going

So to have a much higher chance try for top 16, I personally shoot for 12 (to sleep at end of day 1) as that gets you solidly in as peripheral captain, no brainier second and at worst a high third on alliance 1, 2 or 3... any higher is gravy

Ideally we want top 4. There you can get some of the best at that event then it seems to drop off in most events except for the powerhouse ones.

Re: Alliance Request

Quote:

Originally Posted by Billfred (Post 1566364)

This.

Picture it, Peachtree Regional 2011. 2815 is there with none of our college students (one professor popped an exam on all of them), a pretty slim crew, and a robot that was crated without much in the way of hard driving. It shows on the field Friday, when we die or break something in all but one match. I was the coach (by default, basically) and it was pretty embarrassing coming off of two solid robot years. But we keep working through the different issues, and I share the hard-luck story with a few friends on 2415 and 1771 (the latter of which we were partners with on Friday--their minibot won us a match where we broke).

We got to Saturday morning and a match with clear #1 seed 2415, where they ask if we can play defense. We hadn't had a chance to really work our arm because we kept dying, but we had a three-year driver that knew his stuff. "Suuuure!" We get one clean match off, get the win...and then break in our last match. I'm billed as the resident optimist of any team, but even I knew we had no business being in the playoff rounds. No amount of hard-sell marketing was going to change that.

Yet for reasons I still don't fully understand five years after the fact, 2415 and their first selection 1771 believed me when I said it was a new issue each time, liked working with us behind the glass, and believed our driving and defense was more desirable than the other 24 teams that were available for them to choose even if we had a glass jaw. Six matches later, we're going to St. Louis.

Shoot straight, be positive, and sometimes you'll see your faith rewarded.

There were many reasons we picked 2815. One of the most important aspects in a third partner is "how well do they work with you?". That aspect is often more important than how good the robot is. A prior relationship helps. Remember the private 4-team pits we had at Palmetto in 2009?

Re: Alliance Request

Various thoughts on everything being discussed here:
-I've been in a situation where a team that we didn't want to be picked by came to us and talked about alliances. We were honest, telling them that we believed that there were teams we would benefit from being with more. Honesty is both the most gracious and the most professional thing you can give someone, just know how to deliver it nicely.

-I've noticed teams that are particularly rude or disrespectful. Many of those teams have gone on our "don't pick" list event after event simply because they've gained a reputation for being toxic. These teams don't know how to act professionally, so they don't belong on our alliance.

-We've had tough schedules, it happens. Every team eventually has that match with that team that came in with a kitbot in pieces and asked "what do we do." Speaking from experience, the best thing to do is to get them up and running, and just make sure they know the fundamentals of the game. If you give them simple tasks and a bit of support, you will win matches you didn't think you could possibly win.

-Teams that come to the pit to haggle their way onto picklists are annoying. Don't be that team, you aren't doing yourself any favors. Your strength as an alliance partner is decided by how you perform on the field, how well your drive team works with ours, and how professionally your team as a whole conducts itself.

-In match, don't do things you wouldn't normally do just because you think it'll capture the eye of elite teams' scouters. With the limited depth of analysis that most teams do, that match will just show up as a blip in the data and you'll be written off as inconsistent. Play in a way that is advantageous to your alliance. The only exception I can think of is defense... sometimes. Good scouts can spot a team that consistently plays with good strategy, and those are the kinds of teams that will be picked.

Re: Alliance Request

Instead of building a team dependent robot building a robot that can carry the team solves the "horrible matchup" situation.

Re: Alliance Request

Quote:

Originally Posted by CalTran (Post 1566301)

Perform to the best of your ability. You gotta learn to play the hand you're dealt. If your robot does well, scouting will pick up the rest. There's more to picking a robot than it's win-loss record.

We don't even pay attention to win loss or ranking in our scouting. It's all about how an individual robot team performs. I'll add that having a cooperative attitude goes a long way as that added X factor.

Re: Alliance Request

Quote:

Originally Posted by evanperryg (Post 1568768)

-In match, don't do things you wouldn't normally do just because you think it'll capture the eye of elite teams' scouters. With the limited depth of analysis that most teams do, that match will just show up as a blip in the data and you'll be written off as inconsistent. Play in a way that is advantageous to your alliance. The only exception I can think of is defense... sometimes. Good scouts can spot a team that consistently plays with good strategy, and those are the kinds of teams that will be picked.

I think it is important to understand what type of robots the top seeds will want to pick and showcase those abilities opposed to trying to win a qualification match with a strategy that won't get you picked*. At nearly every event there are a handful of bubble teams that would be better off switching to a support role (defense, feeding, low goal scoring, inbounding (2014)) than showcasing their sup-par scorer. If the top teams are looking for a robot to fill a specific support role and see you do it successfully in a few matches, they will likely forget that went went 1 for 15 shooting in your first 5 matches.

*Obviously follow your alliance's agreed upon strategy...

Re: Alliance Request

Quote:

Originally Posted by Edxu (Post 1566875)

Approach high-seeding teams. Questions that you'll mainly want to ask them run along the lines of "What are you looking for in a second pick alliance partner?". Another great way to communicate teams is to ask them if they want to see anything performance-wise out of your robot in matches.

There is a danger here in fishing for information. We've been to events where clearly someone was fishing for another high seeded team that was their friend (We sort of have do a lot of regionals away from our area, so we are often on the outside looking in). They were then overheard by another of our students saying "sorry, that didn't work. They wouldn't bite." to the aforementioned high seeded team. You run the risk of raising peoples' suspicions if you ask too pointed of a question about their picking strategy.

Know that if you are ranked low and the game requires defense, you should show defense Saturday morning without getting fouls. If you have cheesecaked blockers to add, add them Saturday morning. The high ranked teams will come talk to you if they need you to do something.

Definitely agree with others who have said not to talk to the high ranked teams Saturday before alliance picks. They have more important things to do than satisfy your curiosity. If you are trying to talk to them while they are finalizing their list, you are keeping them from doing what they need to do and making it extremely challenging to be graciously professional. If they are worth anything scoutingwise, they will have been watching pretty closely. I know it's maddening. We have been on both ends of the rankings over the years.

Re: Alliance Request

Quote:

Originally Posted by Caleb Sykes (Post 1566384)

My advice would be to do as much advertising as you can. There are no downsides if you are not annoying about it, and it has the potential to move your team further up other teams' picklists. Our team always advertises ourselves to teams that are higher ranked than us that we would like to be on an alliance with, even if we know we are going to be captains.

Advertising yourself can be a bit awkward at first, but it gets much easier with experience. Here is generally how I approach this process:
First, know what you are trying to get out of advertising yourself. Are you trying to be a first pick? A second pick? A third pick (at champs)? Is the team with whom you are speaking a 1-3 seed and likely captain? A lower seeded likely captain? A lower seeded likely first pick? The way to sell yourself is different for each of these combinations.

When you go to speak with them, start by asking to speak with someone on their scouting team (lead scout is best, but not always readily available). Smaller teams sometimes don't have scouts, in which case you should try to talk with their drive team. If you just talk with some random person in their pit, you are wasting your time. Be polite and respect their time. I have found that teams are generally pretty receptive to talking, but if they are busy or do not wish to speak with you, move on and do not pester them. When you find someone to talk with, make sure to keep their time in mind. If you are talking with a scout in the stands, don't talk during a match, let them work.

Once you find someone to talk with, start by complimenting their team. They have clearly done well or you wouldn't be talking with them. Mention at least one specific things you like about their robot/strategy (I wish we would've thought of..., it was really cool last match when you..., your ... is super slick) but don't spend too much time on this, keep it short and sweet.

Next, and this is really important, you need to tell them things that they might not know just from watching matches. Bring them to the practice field to show them your scaler, and climb 4 times in a row for them. Bring them to your pit and show them the defensive blocker you are building. Tell them that you will be debuting your scoring auto in the upcoming match, so they should keep an eye out for that.

When you have finished with this, ask them if they have any questions for you, and answer them truthfully. I won't speak for others, but I value honesty a lot. If you tell me that your team scores 5 boulders basically every match, and our data says you only scored 3 in your best match, you are likely to go down a notch on our pick list.

After that, ask them what they would like to see out of your team in the upcoming matches, and demonstrate that ability if you have the opportunity. However, make sure you clearly communicate with your partners what you will be doing every match.

One more thing, it is okay to mention that you seeded low, but don't whine to them about your poor schedule and the crappy partners you have been getting, they don't want to hear it, even if it is true. Own up to your own performance. Recognize the failures and limitations of your robot and explain them.

Finally, here are some common situations and key features to sell yourself on in those situations.
1. Selling yourself to a 7 or 8 alliance captain. Tell them truthfully how you did in your best match, and explain exactly what went wrong, and how you have mitigated your issues in your poor matches. Remind them they will be up against the number 1 or 2 seeds, and that they really need someone who has the potential to do really well (you), even if you have been inconsistent previously.
2. Selling yourself to a higher seeded team as a second pick. Emphasize your consistency, higher seeded teams want reliable partners. Show them your defensive blocker that you are building, and stress that your drivers know how to play good defense without getting penalized. Also, stress your ability to score in auto and the endgame.
3. Selling a special/unique ability about your team. Find high-seeded teams that do not also have this ability, and explain how important this ability will be in playoffs. Common examples this year often include crossing A defenses, crossing C defenses solo, climbing, and crossing many defenses in auto.

I agree with you on all points. There is a lot to be said for data and scouting but advertising your team is crucial to be on an ideal alliance. Here are some basic steps for advertising your team.

1.) Initial contact, greeting, and asking to speak with the alliance selector
2.) Flattery, their team is good and there are reasons you want to be with them. Talk about everything they can do well. You should be using scouting data for your claims.
3.) Complement, how do you complement their team, what do you offer to bring the most potential to their alliance. Why should they pick you? What scouting data do you have to show them of your performance? (optional) Do you have video of a recent match in which you did something special like climb? (optional)
4.) Analysis, show off your alliance picking analysis and playoff game analysis to see how things will pan out and what critical ability you will offer. At our 10K regional we knew they weren't many robots who could get in enough balls for their alliance to capture in playoffs so even though we were a low goal shooter we stressed the importance of that capture and how a low goal shooter putting in 4-5 low goals consistently is worth more than high goal shooters who can only score 2 consistently. We didn't always score 4-5 low goals but we focused on matches where we didn't have to be concerned about breaching and showed alliance captains our potential which averages of scouting data might not show. We advertised as a 1st pick.
5.) Overall playoffs strategy, at 10K we saw that the 1st seed alliance had a pretty clear shot to the win (but Highland's alliance was awesome in giving them a run for their money!), however there would be at least 2 wild cards dropped by that alliance. When we went around advertising ourselves we made it clear that our goal was to make it to the finals, pick up those wild cards, and qualify for the Championships. We would try to win if possible, but making it to the finals was our primary, realistic objective. We would achieve this by assuring our alliance the capture and scoring lots of points, using the second pick of our alliance captain to deny the other team the capture because most high goal shooters aren't able to score many boulders under defense (great job 5996!)

We (4th seed) were selected by the Firebears (3rd seed) along with 5996 and we made it to the finals, and our whole alliance is going to Saint Louis! There is a lot to be said for strategy in the alliance selection and advertising your team because quals performance doesn't always reflect the extent of your team's abilities! There's a lot of strategy involved in how you advertise to certain teams as Caleb has pointed out but there's more than I would like to cover in this post. I just wanted to give a framework for the basic advertising plan.

That being said, don't just advertise to every team. That is annoying. Advertise to teams who you want to be on an alliance with and that you have something to offer to that you believe they would like. We didn't advertise to 2052 even though they are one of our favorite teams and are so amazing (congratulations on Chairman's!) because their best options were clearly 525, 5434, or 2502.

Quote:

Originally Posted by IronicDeadBird (Post 1568771)

building a robot that can carry the team solves the "horrible matchup" situation.

If only it were that easy ;)

Re: Alliance Request

Your best "advertisement" is to have a drive team that is courteous, listens and communicates well, doesn't have an abusive drive coach and is well organized. We move teams up our draft list when our drive coach says "we loved working with them." 5012 was a case in point last year for 118 and us. And alternatively most teams have a "do not pick" list built on poor experiences with particular teams.

Re: Alliance Request

Quote:

Originally Posted by Citrus Dad (Post 1572761)

I cannot agree more. Last year on 1257, we had a sheet completed after each match in which the drive rates how well it got along with our alliance partners' drive teams, and stated how much they'd like to work with them in eliminations. We frequently referred to these sheets during picklisting.

Re: Alliance Request

Quote:

Originally Posted by wazateer1 (Post 1572743)

If only it were that easy ;)

Its super easy if you get lucky with the design of your robot the first time every time.
I don't see why people aren't just getting good at being lucky, all you gotta do is make a few sacrifices to the random number generation gods, never walk under ladders, and avoid black cats.

Re: Alliance Request

This is what I made for our team for our 2nd district event. We had a scout go around and hand these out, and handed them to anyone who came by scouting us. Technically, this isn't helping you get picked, but it's better than nothing.