[FVC]: Longer competitions

[rswsmay]

Dave,

We could not find a happy medium when generating the match schedules. When we tried to set a lower minimum match spacing, we sometimes ended up with teams competing in a match while needing to be queued up for another match.

It is possible that we just didn't go through enough iterations to ensure that we didn't have duplication. It also didn't help that we had to generate the match schedule three times because of teams being initially on the list, then not showing up, then finally showing up during opening ceremony. We didn't have enough time to go through the list and check for duplicate matchups. Some of those issues may have contributed to those other events having similar duplicate matchups.

Originally Posted by Dave Flowerday

You need to bring this issue up with the people running the events you attended. The organizers of the event control the amount of duplication seen by teams when they generate the schedule by choosing the "minimum match spacing" (which is the minimum number of matches a team has in between matches they are scheduled to play in). When this number is set too high, duplication will occur (due to basic math constraints). This is documented in the manual.

[gblake]

Originally Posted by rswsmay

Also with new games out each year, FIRST usually has new scoring software written that is game specific.

Correctly written, the score tabulation and related concerns would be an almost trivial part of the collection of routines. Changing the game won't be a big deal. That certainly shouldn't cause a big rewrite.

Originally Posted by rswsmay

I am not sure what to tell you about undertaking something like that. Come September, that reworked code may not be very useful for the newly announced game.

If you want to describe what you like and dislike about the functions and display formats of the current software; I, or the Wildstangs, or whomever will be able to make a better product in the next iteration. If you want to tell me/us that you are "in" then that encourages me to spend some time on it (and would make you the first to cast a vote either way)

Originally Posted by rswsmay

There is also the remote possibility that a new game in the future may utilize 3 teams per alliance (hint hint). The current field controls can accommodate up to 6 teams per field.

Again, if properly written, changing the numbers of teams per alliance (a possibility I and many other folks have noted but don't expect to see until the field is bigger (not supposed to happen soon unless perhaps it is expanded by adding panels to the current perimeter) is a change that also shouldn't be too hard to incorporate.

Blake

[gblake]

Originally Posted by rswsmay

Dave,

We could not find a happy medium when generating the match schedules. When we tried to set a lower minimum match spacing, we sometimes ended up with teams competing in a match while needing to be queued up for another match.

It is possible that we just didn't go through enough iterations to ensure that we didn't have duplication. It also didn't help that we had to generate the match schedule three times because of teams being initially on the list, then not showing up, then finally showing up during opening ceremony. We didn't have enough time to go through the list and check for duplicate matchups. Some of those issues may have contributed to those other events having similar duplicate matchups.

Here is a suggestion for whomever is listening:
Instead of putting the information about match spacing's relationship to duplicated encounters in a manual, and then apparently leaving it up to a harried user's eyeballs and brain to scan the freshly created match listing for "problems"; why not have the software display/print out some statistics (along with each match list it generates) that identify the extent to which those problems actually exist in the matches???

In addition to creating the match list, the software should report any/all things in that match list that a human should assess before that human actually uses the list..... I give myself this sort of help in the Excel spreadsheet that I use.

Blake

[gblake]

Originally Posted by rswsmay

Blake,
...
Hopefully knowing a bit more about the background of FVC as well as FLL will help you understand the comment. ...

None of this is news to me.

Originally Posted by rswsmay

So what I am saying is that as much as you would like to have the learning curve shortened by using ordinary terminology, the fact is that this is pretty much ordinary terminology for the majority of posters here. I know I have read many of your posts regarding programming and the use of sensors and most of that ends up going over my head since my background is Mech. Engineering.

OK, however, the point is that those messages (hopefully) should not go over the head of any FIRST or non-FIRST person who has a background in the subject. The key thing is that I try not to assume that readers are "in the FIRST club". When thousands of new FVC teams appear over the next few years, anyone who doesn't make their communication attempts "FIRST-neutral" will fail to communicate.

I don't want to shorten the learning curve for me, I am already pretty far along it. I want to point out (assert) that much, perhaps most, of the "learning curve", in the sense that I am trying to convey, is an unnecessary stumbling block that will impede FIRST's progress. Carried to an outrageous extreme, clique-ish-ness within FIRST could result in the organization being a fading set of early adopters that are being absorbed into some newly-minted, 10,00-team organization rather than actually being that 10,000 team organization.

Originally Posted by rswsmay

Nonetheless, I would expect that the majority of posters that read through your posts find them informative, useful and totally appropriate including me at times.

Thank you, I try.

Originally Posted by rswsmay

Your point is well taken though… so when appropriate I will refrain from using references to FLL without laying some background.

Remember when I wrote this above: "This comment ... is a ... general one to folks discussing FVC." Along those lines I suggest that every volunteer involved in FIRST today (FRC has penetrated less than 5% of US high schools) needs to think about how our roles might change (voluntarily or inexorably) if we were suddenly in a sea of 10 times as many teams (FRC + FVC). I suggest we all start immediately training our several lieutenants/replacements now, in order to get ready for tomorrow....

Today, when I look around, I notice that many vital volunteers accomplish their goals through laudable, heroic effort; but that won't cut it if FVC takes hold and raises FIRST market penetration from under 5% to somewhere around 50%. The smarter not harder cliché comes to mind.

Now if only I were smart......

Blake

[Cartwright]

Originally Posted by MGoelz

Especially with VEX being in only its 2nd official year, teams need more time to gain more experience. I participated in VEX during the pilot season and we went to St. Louis. I think, at the time, that was one of the nearest competitions. I think we just need to let it grow a bit, expand competition sites, and see where it goes from there. Most of it is going to depend on participation, and what seems to fit well with this program.

VEX does indeed need to grow a bit more. I don't think people recognize its potential. Our team has a satellite VEX team at Perry Meridian Middle School that I was a part last year, and while there are many highschool VEX teams, I think we found that the middle school age group is more than capable especially with the mentoring of high school students. Labeling VEX as middle school robotics could help bridge the gap between FLL and FRC. On a scale, FVC I think should be between FLL and FRC, not closer to FRC as I see it now.

[gblake]

Originally Posted by GaryV1188

I can see FVC expanding greatly in the next few years. Probably never as big as FLL, but maybe halfway between FRC and FLL. We will have to have hundreds of VEX tournaments to handle the number of teams. That's either a very scary or a very inspiring thought!

Originally Posted by fredliu168

That would mean FVC would have to increase the number of tournaments teams are allowed to win awards in. Also, it would probably mean more divisions in the championships, like the FRC 4 divisions.

I actually love that idea.

Originally Posted by GaryV1188

More likely, it will mean that FVC gets 3 tiers of competitions like FLL has now. You win(*) at a regional, and you advance to the next level. You win(*) at that level and you qualify for the Championships. Currently in the FLL model, the "next level" is state or country. For FVC, it would be a larger section of the US or a few countries for international teams.

Unfortunately, there would be travel and other costs associated with the 2nd tier tournaments, meaning the path to Atlanta (or wherever) just got more expensive.

(*) - "Win" wouldn't necessarily mean just the first place alliance and the Inspire award. Assuming maybe 20 second-tier tournaments of about 40 teams each around the world, 6-8 teams could advance from each tournament to Atlanta, and compete in a manageable-sized competition (with divisions!) of 120-160 teams. Maybe first and second placed alliances, Inspire plus runner up, etc. could advance. The number of teams advancing from the regional to the 2nd tier would also be determined in advance, based on number of entries. For example in Michigan FLL, about 1/4 of the teams entered in the regionals advance to one of the two the State Tournaments. This structure could support 3000 VEX teams, less if they are allowed to enter more than one regional.

Trying to get back closer to the original intent of this thread, I have been trying to think of competition models that are employed by other organizations; that might also work for an FVC season enjoyed by thousands of teams. Key points seem to be these:

• Many other competitions/challenges that are both spectator sports and that place participants in head-to-head competition, typically have a long regular season filled with matches that give all participants a good dose of time "on the field". FVC doesn't have that but tries to give a worthwhile experience to teams in 1-day tournaments (Petty hard to do with a large number of teams and a small number of layers of competition)
• Other competitions do not muddy up a team's chances to prove themselves worthy of advancing, by allying them with pseudo-randomly chosen partners in a handful of matches on a single day. FVC does this, but could instead switch to other formats. Many are possible. Most or all the (1-day) alternatives would poke a great big hole in the concept of coopetition. Most or all of the whole-season options would raise the specter of high costs, if the team-density didn't rise to something like the densities other popular competitions in this age group enjoy.

So, if FVC grows to a few thousand teams, if the number of layers of of expensive competitions between a team and the World Championships is kept low, if coopetition concepts force teams to ally with random partners during tournaments, if doing well at 1 or 2 "championships" per season remains necessary for getting to the World Championships, and if the FVC Tx/RX technology doesn't change to one that can use ID codes to share RF channels across multiple fields; then I think that multi-day competitions are going to be necessary in order to give the large number of tournament participants at each event a statistically significant number of matches and enough matches to make the season worth their investment.

Change a few of the "if"s above and maybe two-day events won't become necessary.

Blake

[ManicMechanic]

Originally Posted by 1885.Blake

[*]Many other competitions/challenges that are both spectator sports and that place participants in head-to-head competition, typically have a long regular season filled with matches that give all participants a good dose of time "on the field". FVC doesn't have that but tries to give a worthwhile experience to teams in 1-day tournaments (Petty hard to do with a large number of teams and a small number of layers of competition)

At the beginning of this season we attempted to form a somewhat formalized league with "league play." Basically, a league would consist of 8 - 12 teams, each of which would play 2-4 events/season, then participate in the Championship in the Spring. Not a single one of the potential dozen teams we contacted was interested. We cut the number of required events from 4 to 2 to 1, but no bites. Money was not an issue for these teams -- 20+ kits were purchased for their use, their registration & tournament fees were covered, and even the travel expenses to Atlanta were promised for one team if they qualified. But the teams were so intimidated by the perceived time commitment that none of them even came to the Championship.

Far more successful were the 2 scrimmages held by the Northern CA State partners. These events were "come as you are." Some teams came with reasonably advanced robots, others came with never-opened Vex kits in boxes. At the first scrimmage, teams mainly built and tested on the field throughout the day (several teams had no practice field or elements at home), and around 3:00, a series of 2 vs. 2 matches were held, open to any team that had something that rolled on the ground. Because of the small number of teams (8), each team had the chance to play 6 or 7 matches (run in rapid succession over the course of an hour). My team loved it because every member got to driving practice, something that would never happen at a formal event.

The more that is required by an event (time, money), the harder it is to get rookies to come. Once they attend their first event, they typically become "hooked" as they are inspired by those around them. At that point, having more complicated events (2-day, or multi-event leagues) is very practical and desirable. But the trick is to get them to the starting gate.

I think that having smaller regionals (about 20-ish teams) that send the top-performing teams to larger Championships could accomplish this goal. With fewer teams, you can run more matches, and scouting is much easier. The "winners" are more willing to spend the money to attend a 2-day Championship, because they feel pride at having "earned" their place, and the excitement of their success at the first event provides sufficient momentum to propel them to the next level.

While some don't like the idea of having 3 layers to get to Atlanta, the reality is that in the future, Atlanta may become a privilege enjoyed only by a very select few -- for example, in FLL only 100/8000, or < 2% of teams attend the World Festival (vs. 100/550 ~18% of FVC teams). At some point, advancing to a Championship may be considered almost as much of an honor as advancing to Atlanta is now, but with less travel.

[gblake]

Originally Posted by ManicMechanic

...
Far more successful were the 2 scrimmages held by the Northern CA State partners. These events were "come as you are." Some teams came with reasonably advanced robots, others came with never-opened Vex kits in boxes. At the first scrimmage, teams mainly built and tested on the field throughout the day (several teams had no practice field or elements at home), and around 3:00, a series of 2 vs. 2 matches were held, open to any team that had something that rolled on the ground. Because of the small number of teams (8), each team had the chance to play 6 or 7 matches (run in rapid succession over the course of an hour). My team loved it because every member got to driving practice, something that would never happen at a formal event.
...

I have to agree, even though it might appear to conflict with what I said above. I think the "bring anything that rolls" model is one to emulate when bootstrapping new teams.

What has evolved into the community (still very loosely organized) that is becoming the Potomac Vex League, started with a couple of come-as-you are "practice session" and every month since then has attempted to hold a workshop or scrimmage if something big isn't already occurring in the region that month.

Giving teams, school classes/clubs, etc. the early chances to get their machines built and driven in nearby, low-key, no-risk, supportive somethings called practices, workshops, and/or a come-as-you-are day of fun, seems to be an effective way to get them over the initial hump. Then after they have played a little metaphorical tee-ball, they are ready to ease into a steadier diet of periodic events.

Next season I hope to see 5 or 6 new formally registered teams come out of this year's informal league regulars, and see the league pipeline become primed by another 5 or 6 groups who are repeating the process we seemed to get right this year (or we just got lucky - time will tell).

To connect this to my previous message and to MM's comments: Announcing to a group of complete newbies (perhaps containing a high percentage of NEMOs) that a hard-fought competitive season is going to occur if they sign up, might scare them away. Easing them into a steady diet of periodic competitions can work well; and then once they are in the habit of inspiring students by steadily giving the sustained positive feedback (fun) of frequent competitions, FIRST can enjoy the side-effect of not having to make every FIRST championship so long and exciting that it makes the entire season of work worthwhile. The league play took care of much of that.

However, there is still the matter of figuring out a way to run a championship's field-competition so that the participants generally agree that the best teams have bubbled to the top. Giving teams plenty of time on a league field before the championship only hits some of the important aspects of a successful FVC program, not all of them. Ensuring that enough championship matches are played in a format in which the luck-of-the-draw is substantially less important than the skill of the team, is not solved by league play unless, the league play results are somehow factored into the championship matches.

Summary:
Come-as-you-are practice sessions & workshops = Good for each/every early season and especially for new clubs/teams/classes.
Frequent league events = Good for helping teams feel they get enough return on the time and money they invest during a season.
How to best organize champoinship matches = ???, but a small handful of Quals doesn't seem to sit well. There is too much random luck involved.

ManicMechanic - Did I get this summary right?

Blake

[ManicMechanic]

Originally Posted by 1885.Blake

Summary:
Come-as-you-are practice sessions & workshops = Good for each/every early season and especially for new clubs/teams/classes.
Frequent league events = Good for helping teams feel they get enough return on the time and money they invest during a season.
How to best organize champoinship matches = ???, but a small handful of Quals doesn't seem to sit well. There is too much random luck involved.

ManicMechanic - Did I get this summary right?

Blake

Yep, that about sums it up. One thing I might add about qualifying matches at Championships is that although the number of matches matters, it won't do any good if the DIVERSITY of matches doesn't improve. For example, we had 2 matches where we had exactly the same alliance partner and same 2 opponents. Big surprise: the scores of the matches were very similar -->wasted extra match. While I now understand that the software & minimum spacing were at play here, there's no reason why the same 4 teams couldn't at least have the alliances permuted differently, i.e., Teams 1 & 2 vs. 3 & 4 in one match, then 1 & 3 vs. 2 & 4 next match, rather than 1 & 2 vs. 3 & 4 in both matches. At the very least, the software needs to have a manual override so that matches like this can be rearranged. But really, there needs to be some sophistication added to the software that deliberately avoids duplicates wherever possible. While I can do the math/algorithm, the coding is beyond me, and as is becoming my habit, I am leaving the hard work to some unwitting soul who is more competent than I.

[gblake]

Originally Posted by ManicMechanic

While I can do the math/algorithm, the coding is beyond me, and as is becoming my habit, I am leaving the hard work to some unwitting soul who is more competent than I.

You send me an algorithm suitable for implementing in software; and I'll both write it up in open source, and wrap a user interface and some output formatting around it.

I need to do this for myself anyway for scrimmages, having some help will make life easier, not harder. I have been nibbling around the edges of the problem; but I haven't hit upon an approach that I like yet. The result has been two long nights of manually building lists of match pairings (for N=20 and for N=16) (Ugh).

Blake

[billw]

The randomness as well as the ranking point system bother me as well, and I am not sure it delivers the expected results. But that is another discussion.

Would it be an improvement if teams were sub-grouped into lots of four and then had to play three matches between themselves? Because everyone would be on the field, it might be much faster to run. You might even be able to repeat the regrouping three times (each team would then play nine matches).

Without giving this more than 20 seconds thought, I think this might reduce the effect of random partners causing such wild swings (more matches certainly would).

--Bill Wiley

[ManicMechanic]

Originally Posted by billw

The randomness as well as the ranking point system bother me as well, and I am not sure it delivers the expected results. But that is another discussion.

One possible improvement to the point system is to have rank points based not on the losing alliance score alone, but finding the differential between the winning & losing scores and adding this amount to the winning alliance's rank points, and subtracting the same number of points (ouch!) for the losing alliance. That way, a strong alliance is not penalized by having low-scoring opponents.

On the other hand, perhaps FIRST is trying to add some uncertainty to the strategy of the game -- an alliance that sees that its opponent is low-scoring might strategize to help its opponent score some points near the end, benefiting them both. I suspect this is why 3053 didn't try for the atlas ball in a 1 vs. 1 match we played with them.

Originally Posted by billw

Would it be an improvement if teams were sub-grouped into lots of four and then had to play three matches between themselves? Because everyone would be on the field, it might be much faster to run. You might even be able to repeat the regrouping three times (each team would then play nine matches).

Yes, it's faster to run, but I think there are limitations to this situation. Suppose that 4 of the best teams are grouped together. Half of the outcomes must be losses, whereas all 4 teams might have had all or or mostly all wins had they played some weaker teams. Conversely, in a grouping of 4 weak teams, at least 1, possibly 2 teams could have all wins, and look like an artificially strong team. Ideally, you want to see each team play with and against a variety of strong, intermediate, and weak teams. This also allows a team to show off the range of its abilities for scouting purposes. However, your idea has spawned a model I'm developing where clusters play in groups of 8.

The 2 models that have invaded my brain this past week look like the following:

Increment Model - For each match, separate team numbers by different increments.
Advantages:
• No duplicate alliance partners
• Minimal duplication of opponents
• Odd number of teams is easily dealt with

Disadvantages:
• No provision made for minimum spacing

Node Model - Form clusters of 8. Have each team play 7 matches (or fewer) against the other teams in the cluster.
Advantages:
• No duplicate alliance partners
• For tournaments of > 15 teams, minimum spacing of at least 2 matches
Disadvantages:
• Duplicate opponents are likely
• For number of teams not divisible by 8, special provisions must be made (if the remainder is close to 8, leave gaps and fill in with volunteers. If the remainder is close to 1, mix in the extra teams, extending the cycle length by 1 match.

The details take 9 pages. and are attached. If you can see improvements or ways to overcome the disadvantages, please send them along.

[ManicMechanic]

Originally Posted by ManicMechanic

The details take 9 pages. and are attached. If you can see improvements or ways to overcome the disadvantages, please send them along.

Oops, thought I'd attached but found out it was too large. It's posted as a White Paper here:
http://www.chiefdelphi.com/media/papers/1996

Yolande

[Ben Mitchell]

Originally Posted by Rich Kressly

Good discussion.

From a programmatic standpoint, FVC has been committed to being more affordable and accessible up to this point. If that holds true in the future, I wouldn't expect to see multiple day regional events soon. Missed school and overnight travel, while a foregone conclusion for some, is a deal-breaker for many underserved, inner-city, and rural schools and teams that have scarce resources.

If I had to take an educated guess now, I'd be putting my energy into how to improve on the one-day format.

The one day format I feel worked great for the smaller tournaments.

I personally would rather see more numerous smaller events than a fewer number of larger events.

Really its an issue of limiting the number of teams in order to give the teams at the event a reasonable number of matches.

That being said, FVC has had this year a quick match-to-match turnaround (2 fields is a must) so we were able to play more matches much faster than FVC in event.

At this point, I would very much agree with Rich - Keep it to one day. With more events held with fewer teams, multiday events arent needed.

[KathieK]

Originally Posted by Ben Mitchell

At this point, I would very much agree with Rich - Keep it to one day. With more events held with fewer teams, multiday events arent needed.

More events with fewer teams = more tournament sponsors needed to help underwrite the costs of holding the events.