Make Scounting Count: NFL Draft Style

As a true FIRST addict, my dad and I were discussing different ways to do the alliance pairing at competitions, and that was when we thought of the NFL draft.

If FIRST used a similar system, it would encourage teams to scout the other teams out on their field, and not just choose favorites. (Admit it… wasn’t at least one of your alliance partners a “friend” of your team?)

Last weekend at the championship, for my first time since joining the team in 2003, SPAM was not chosen for the finals after lunch. This year, our team was scouting in the stands, but we did not do much “advertising” around the pit area. Most people saw our robot in the pits, but did not have the opportunity to see her on the field… this I feel, could have changed our outlook.

Just think… if teams had to submit their list of desired alliance partners, and FIRST had a computer system of organizing their selections, similar to the NFL draft (Letting the top 8 seeded teams select their alliances this way, even allowing them to select one another, and pushing the 9th up…there would be no more “buddy system”) The unknown teams would have a better chance, because scouting would be taken more seriously… who has the best bot on the field?

I know there are flaws to this system, (like, well, we’ve already got a ramp bot in our alliance) but, there is potential. Just wondering what the Chief Delphi Community has to say about this concept. :slight_smile:

Im not sure I am understanding your idea…can you explain what happens by “submitting” teams ?

I think this is a great idea, but that’s just my opinion. I’m sure there is a reason why FIRST chooses to do the drafting process like they do. No matter what, there will always be some controversy though. We are all robotics people, and we are always looking for something to invent or improve. Therefore, to me, this idea sounds like it would work rather smoothly, however I would still like to see what everyone else thinks…:slight_smile:

Also, I think teams become “friends” with each other by drafting each other. So, eventually you should have a ton of “friends,” right?:smiley:

your team would have to scout, and submit a list of your top ranked teams (that you would like to be aligned with.) The computer would take the top ranked team’s 1st choice, align them…then make the 2nd ranked team’s 1st choice, and so forth, for the top 8 teams, then make a 2nd round for their next pick. “Drafts” (chosen alliance partners from the pool) can be traded among teams during a short time period before the final alliances are called.

This would make it more important to scout the best bots on the field.

Isn’t this basically what FIRST already does?

Are you talking about the top teams all submitting lists then a computer making the picks according to the rankings on those lists?

First, I’m not sure the NFL (or any professional sports) work that way. In the NFL draft, teams are given 15 minutes to make their selection. I’m pretty sure team don’t submit lists and the NFL then distribute the players.

I don’t think a system like this would save any time, rather it would frustrate teams. Many picks can be situational. For instance, we had a list of 16 or so teams that were ranked. We then had a second list of teams (3 or 4) that if they were available for the second pick, we’d take them, if not we’d default back to the first list. I’m not sure how a computer would handle a list like that.

Many times, while alliance selection is going on, teams are keeping track in the stands, then signaling to their representative who to pick. The second round pick could change depending on who the alliance will be facing, who compliments the first round pick best, and other factors. If I select team ABC in the first round I may want team DEF in the second. But, if I get team HIJ instead of ABC I may want team KLM instead of DEF. (sorry for all the letters)

I think if a computer selection system were implemented, you’d hear a lot of “Oh man, team XYZ was available for our second pick? They wold have been better than team UVW that we got.”

Personally this is a BAD idea. I don’t mean to insult you or anything but to be honest, strategy changes depending on who picks who.

Sometimes depending on who picks who before you, you need to change who you will pick, to counter such a team. Scouting isn’t always about who is the best, sometimes it is about which teams can beat which teams.

For Example:

Team A is a really good tube scorer and doubel ramper
Team B is a really good tube scorer and single ramper
Team C is a Great tube scorer
Team D is a Great tube scorer
Team E is a great defender and a double ramper
Team F is an ok scorer and a great double ramper
Team G is a ok scorer and an ok double ramper and a great defender

Now lets assume

Team C - #1 Seed
Team D - #2 Seed

Team C selects Team A - Team C & A
Team D selects Team F - Team D & F

Now the other teams not selected are left on the way back.

Team D & F select Team G - Team D F G
Team C & A select Team B - Team C A B

Now if Team C had selected differently in the first round this could have been what happened:

C picks E
D picks A
D & A pick B - Team D A B
C & E pick G - Team C E G

Now team E was picked this time around and Team F was left out. The reason for this is that Team C was looking for something different, so Team D changed there strategy part way through the selection process.

Just my 2 cents

P.S. if you don’t understand this just PM me and i can try to explain a little better

I think there would be a few flaws to the system. For this year, the ideal alliance (in my eyes) was 2 strong offensive robots and a ramp bot (or other lift style bot).

There would have to be at least 2 classes of robots needed, offensive and ramp, and maybe a 3rd for defense. Anyway, a team may want that type of configuration no matter what (2 offensive and 1 ramp). If there weren’t 2 categories of robots, things could get messed up, such as automatically assigning 3 strong offensive robots to an alliance or 1 offensive robot and 2 ramp robots.

Thats why he said there is a short trade process, read above.

I like the idea, but it takes away the idea of declining.

i’ll give you two examples

2006 Florida regional

233 was ranked first

they selected 357 (i think) to be their first selection, they declined, being ranked 5th (eventually beat the #1 alliance in the semi’s)

next they went after the second ranked team 86, 86 declined and a pouring oh “ohh’s” came down (note most of these kids, well the kids on my team thought pink was some sort of a bully, and people sticking it in their face amazed them somewhat.)

and last week the curie alliance selections, 1114 and 330 turned down 1732

you wouldn’t get these dramatic stories if you did it by computer, i like it the way it is, our team became friends with 433 and 343 this way, it worked out great.