Toughest Regional Metrics...

[Joe Johnson]

Those of you who attended the Great Lakes Regional and the Western Michigan Regional, know that Dave Verbrugge -- MC without peer -- pulled together a metric that he used to demonstrate (at least to the folks listening to him live) that the GLR was the toughest regional and the WMR was the toughest "pound for pound."

His metric was essentially this: Take all the number of teams that were finalist or winners of the divisions in Atlanta and see which regional had the most number of teams in that group of 24.

If you do that this year, this is what you get:


West Michigan Regional, Allendale, MI 6
Sacramento Regional, Davis, CA 5
Great Lakes Regional, Ypsilanti, MI 5
Detroit Regional, Detroit, MI 5
Finger Lakes Regional, Rochester, NY 3
Boilermaker Regional, West Lafayette, IN 3
Florida Regional, Orlando, FL 2
Pacific Northwest Regional, Portland, OR 2
Buckeye Regional, Cleveland, OH 2
Midwest Regional, Chicago, IL 2
Silicon Valley Regional , San Jose, CA 2
Palmetto Regional, Columbia, SC 2
BAE SYSTEMS Granite State Regional, Manchester, NH 1
NASA / VCU Regional, Richmond, VA 1
Arizona Regional, Phoenix, AZ 1
Pittsburgh Regional, Pittsburgh, PA 1
St. Louis Regional, St. Charles, MO 1
Chesapeake Regional, Annapolis, MD 1
New Jersey Regional, Trenton, NJ 1
Colorado Regional, Denver, CO 1
Las Vegas Regional, Las Vegas, NV 1
Southern California Regional, Los Angeles, CA 1
Peachtree Regional, Duluth, GA 0
UTC New England Regional, Hartford, CT 0
SBPLI Long Island Regional, Hempstead, NY 0
New York City Regional, New York, NY 0
Philadelphia Regional, Philadelphia, PA 0
Waterloo Regional, Waterloo, ON Canada 0
Greater Toronto Regional, Mississauga, ON
Canada 0 Lone Star Regional, Houston, TX 0



Looks good for WMR. With SAC, GLR and Detroit right behind.

But what if you take the WINNERS of the divisions:


Sacramento Regional, Davis, CA 5
Detroit Regional, Detroit, MI 4
West Michigan Regional, Allendale, MI 3
Great Lakes Regional, Ypsilanti, MI 3
Silicon Valley Regional , San Jose, CA 2
Finger Lakes Regional, Rochester, NY 1
Florida Regional, Orlando, FL 1
Buckeye Regional, Cleveland, OH 1
Midwest Regional, Chicago, IL 1
BAE SYSTEMS Granite State Regional, Manchester, NH 1
Arizona Regional, Phoenix, AZ 1
Chesapeake Regional, Annapolis, MD 1
New Jersey Regional, Trenton, NJ 1
Colorado Regional, Denver, CO 1
Las Vegas Regional, Las Vegas, NV 1
Southern California Regional, Los Angeles, CA 1


Sacramento is looking pretty tough.

But you can go with the FINALIST AND CHAMPIONS only



Sacramento Regional, Davis, CA 3
Great Lakes Regional, Ypsilanti, MI 2
Detroit Regional, Detroit, MI 1
West Michigan Regional, Allendale, MI 1
Silicon Valley Regional , San Jose, CA 1
Buckeye Regional, Cleveland, OH 1
Arizona Regional, Phoenix, AZ 1
New Jersey Regional, Trenton, NJ 1
Colorado Regional, Denver, CO 1
Las Vegas Regional, Las Vegas, NV 1
Southern California Regional, Los Angeles, CA 1



Again, Sacramento is looking pretty tough.

But WAIT, THERE'S MORE! What about the CHAMPIONS ONLY:



Great Lakes Regional, Ypsilanti, MI 2
Sacramento Regional, Davis, CA 1
Detroit Regional, Detroit, MI 1
West Michigan Regional, Allendale, MI 1
Buckeye Regional, Cleveland, OH 1
Southern California Regional, Los Angeles, CA 1



Bottom line: Define your metric and you can get just about any result you want.

So... ...what is your metric? Define a metric and argue that this or that regional is the biggest and baddest going on.

Joe J.

[Bill Moore]

Those of you who attended the Great Lakes Regional and the Western Michigan Regional, know that Dave Verbrugge -- MC without peer -- pulled together a metric that he used to demonstrate (at least to the folks listening to him live) that the GLR was the toughest regional and the WMR was the toughest "pound for pound."

His metric was essentially this: Take all the number of teams that were finalist or winners of the divisions in Atlanta and see which regional had the most number of teams in that group of 24.

If you do that this year, this is what you get:


Looks good for WMR. With SAC, GLR and Detroit right behind.

But what if you take the WINNERS of the divisions:


Sacramento is looking pretty tough.

But you can go with the FINALIST AND CHAMPIONS only



Again, Sacramento is looking pretty tough.

But WAIT, THERE'S MORE! What about the CHAMPIONS ONLY:



Bottom line: Define your metric and you can get just about any result you want.

So... ...what is your metric? Define a metric and argue that this or that regional is the biggest and baddest going on.

Joe J.
Wow!

I could have sworn that Robbe Extreme (56) Championship Finalists were winners at Philadelphia. They sure fooled me accepting that trophy there!

[Joe Johnson]

Wow!

I could have sworn that Robbe Extreme (56) Championship Finalists were winners at Philadelphia. They sure fooled me accepting that trophy there!

My mistake.

Others feel free to check my data.

But the challenge is still out there: Propose a Metric for the toughest regional, make it as simple as "this is the one MyTeam goes to" to one that requires multivariable calculus.

Share your thoughts.

Joe J.

[EricH]

How about adding in a factor for number of teams at a regional? (In which case, any double-field regional may have an edge over the single-field regionals.)

[Joe Johnson]

How about adding in a factor for number of teams at a regional? (In which case, any double-field regional may have an edge over the single-field regionals.)

Are you arguing that an 80 team regional that has one Divisional Champion &/or Finalists harder than a 40 team regional that has one Divisional Champ &/or Finalist?

Again, this is a create your own metric thread, but if anything I would argue the othe way: A regional seems harder to me if they have a higher percentage of Divisional Champs & /or Finalist.

So I would argue for the Metric that Dave Verbrugge used at Western Michigan: Pound for Pound = Use percentage of teams that meet the standard at a regional.

I am not sure how big Sacramento, but Detroit was pretty small. I think that Detroit might have a case for the strongest regional on a pound for pound basis.

Joe J.

[EricH]

Are you arguing that an 80 team regional that has one Divisional Champion &/or Finalists harder than a 40 team regional that has one Divisional Champ &/or Finalist?

Again, this is a create your own metric thread, but if anything I would argue the othe way: A regional seems harder to me if they have a higher percentage of Divisional Champs & /or Finalist.

I am not sure how big Sacramento, but Detroit was pretty small. I think that Detroit might have a case for the strongest regional on a pound for pound basis.

Joe J.

That was not my intent, but it seems that number of teams should be a factor, though maybe a small one.

You're right, a regional is probably harder if it has a larger percentage of divisional champion/finalist teams. However, one other factor is: do the top dozen or so teams from the regional go to championships? LA only had three of its top teams at Championships: 980, 22, 330. The other 3 regional winner/finalist teams (69, 634, 968) and all of the semifinalists were not there.

Someone said that you could get whatever result you wanted by manipulating the data just right. They are correct.

P. S. Sacramento had about 36 teams, at least 5 of which made it to Einstein.

[RyanMcE]

It seems to me that the best way to do this would be to look at teams that attended multiple regionals and the championship, and see how each they did in each regional. Specifically, I would like to make a compuer program that uses an algorithm to calculate relative regional stengths using this sort of system. Unfortunately, so much of the information on FIRST's website is missing or completely wrong, I don't think this is possible this year.

But I will offer this: If you are not allied with team 254, the Sacramento regional is the toughest regional to win in the world.

[Winged Globe]

But I will offer this: If you are not allied with team 254, the Sacramento regional is the toughest regional to win in the world.

Or perhaps Silicon Valley? That's where their undefeated streak is. It'll be an interesting and exciting set of matches if/when that streak ends.

[JohnnyB]

Looks like Michigan is representin' *raises roof*

Pretty cool

[Bill Moore]

I don't think you can use metrics from Championships to determine the toughest regionals, because the game and robots change throughout the season.

One could argue that the regionals held the first weekend are the toughest. This isn't baseball or football which is constant from year to year, this game is "newly invented" each year. The teams competing during the first weekend have no previous games to watch for strategy or how opponents may try to disrupt your robot. They have to adjust on the spot to unforeseen challenges, while the rest of us get to watch a broadcast/webcast and learn from their efforts.

Likewise, teams get to view what is important in the playoffs. How many times did you see aggressive defense in the regionals. A number of teams saw penalties because of this aggressiveness. On Einstein, how many robots played aggressive defense for their alliance?

The game starts, we learn, the game changes, we learn some more, the game changes again, we learn even more . . .

[E. Wood]

If a regional has 35 good robots out of 40 and only one national finalist does that make it easier than a regional that happens to have a good robot that happened to be the National Champion? Now this is just an example but regionals that have more team are usually harder than smaller ones just because there are more good teams. Also has the weeks go by some regionals get harder because of the experience that some teams have gained at other regionals. I went to the Pittsburgh regional as a spectator and the Chesapeake regional as a participate and i can tell you that from what i saw it was a lot harder to get into the final 8 alliances at the Chesapeake regional. I am not saying that the winners of the Pittsburgh regional weren't as good, they obvious did well at nationals because Pittsburgh was ranked higher than the Chesapeake regional. But you just cant use the winners of nationals as a way to rank the regionals because not every good team gets to go to nationals.

[Cory]

Joe: it seems to me that you've fairly successfuly debunked the notion that the west coast is "soft" and cannot compete with the midwest/east coast :) (at least this year, and yes, I know, 56 did come cross country and helped our numbers out)

[Rob27]

One could argue that the regionals held the first weekend are the toughest. This isn't baseball or football which is constant from year to year, this game is "newly invented" each year. The teams competing during the first weekend have no previous games to watch for strategy or how opponents may try to disrupt your robot. They have to adjust on the spot to unforeseen challenges, while the rest of us get to watch a broadcast/webcast and learn from their efforts.

You can argue the opposite, too. By the last weekend of regionals, teams have learned the game, and you have to face the teams that have seen the finals and know how to play them.

Take West Michigan (My pick for the toughest regional) for example.

Out of 41 Teams -
There were 2 previous two-time champions: 67, 245
2 previous champions: 66 (who became a double champ), 494
5 previous finalists: 33, 93, 123, 288, 322
the first regional for only 1 team: 518
and the third regional for 7 teams: 33, 47, 67, 141, 245, 302, 494

To me, it's harder to play a team who knows what they're doing than one who is trying to figure out the game.

[Paul H]

the first regional for only 1 team: 518

It was 1504's first regional too.

[Mark Garver]

I don't think you can use metrics from Championships to determine the toughest regionals, because the game and robots change throughout the season.

The same is true because each of the divisions where "stacked" differently which wouldn't allow some teams to surface as others may have in different divisions. The whole use of the bracket system really doesn't inform people to more than who won and who lost. Trying to compare even quarter finalist from the same division, in order to rank is not possible in less some type of "losers" bracket is created where all alliances play each other; however this even assumes that the robots will be of the same quality at the beginning of each match.

Lots could be done to improve the method in which teams advance...

[Kris Verdeyen]

How about average score?

In a year where the game had more of a defensive bent, that might not have been as good a metric, but this year it probably was.

[Jeremy.245]

It seems to me that the best way to do this would be to look at teams that attended multiple regionals and the championship, and see how each they did in each regional. Specifically, I would like to make a compuer program that uses an algorithm to calculate relative regional stengths using this sort of system. Unfortunately, so much of the information on FIRST's website is missing or completely wrong, I don't think this is possible this year.

But I will offer this: If you are not allied with team 254, the Sacramento regional is the toughest regional to win in the world.

Well we weren't allied with 254 and we won Sacramento...YAY for 245. we are thinking of going to silicon valley next year to either break 254'S streak or win it with them

[whakojacko]

Well we weren't allied with 254 and we won Sacramento...YAY for 245. we are thinking of going to silicon valley next year to either break 254'S streak or win it with them

That would be fun, we had a great time with you guys and 766 at sacto :). Im happy to see a west coast regional ranked so "high" on the list, especially with a relatively few number of teams attending.

[Meyerman]

But I will offer this: If you are not allied with team 254, the Sacramento regional is the toughest regional to win in the world.

didnt the poofs only win sacramento once,im pretty sure svr is the one they havnt lost. but sacramento was a good regional team 56 enjoyed the competition i dont know if we will be goin back but theirs always a chance.

[Cory]

didnt the poofs only win sacramento once,im pretty sure svr is the one they havnt lost. but sacramento was a good regional team 56 enjoyed the competition i dont know if we will be goin back but theirs always a chance.

Poofs lost Sac in 2003 (finalists) won in 2004 lost in 2005 (finalists)

One of only three regionals they've lost since 1999

[UlTiMaTeP]

Those of you who attended the Great Lakes Regional and the Western Michigan Regional, know that Dave Verbrugge -- MC without peer -- pulled together a metric that he used to demonstrate (at least to the folks listening to him live) that the GLR was the toughest regional and the WMR was the toughest "pound for pound."

His metric was essentially this: Take all the number of teams that were finalist or winners of the divisions in Atlanta and see which regional had the most number of teams in that group of 24.

If you do that this year, this is what you get:


Looks good for WMR. With SAC, GLR and Detroit right behind.

But what if you take the WINNERS of the divisions:


Sacramento is looking pretty tough.

But you can go with the FINALIST AND CHAMPIONS only



Again, Sacramento is looking pretty tough.

But WAIT, THERE'S MORE! What about the CHAMPIONS ONLY:



Bottom line: Define your metric and you can get just about any result you want.

So... ...what is your metric? Define a metric and argue that this or that regional is the biggest and baddest going on.

Joe J.

The only thing I will say, is the fact that many teams did go to 2 or 3 regionals this year, and from what it looks like, that was not taken into account.

[MFS766]

Well we weren't allied with 254 and we won Sacramento...YAY for 245. we are thinking of going to silicon valley next year to either break 254'S streak or win it with them
That will be a lot of fun if you guys come. In 2002, we lost the finals of the post-season Cal Games to 254. We have since sworn to defeat them at all opportunities :D . I took us 3 more years to achieve our goal at Sacramento. We have since sworn to break 254's streak at Silicon Valley. I hope 245 will come again.

Now, some think Sacramento seems to be really tough with amazing robots. We at 766 found it easy because we practiced with our robot a lot before we shipped it. So when we were at Sacramento, there was little issues that needed to be corrected. The vast majority of robots at Sacramento were driving for the first time, tinkering with their robots for the first time with an arena. Thankfully, though ironic, 254 allowed us to use their practice arena before the ship date. So robots that had significant test times dominated. Examples are 245, 254 (they had two robots built), 56, 330, 1097, 1072 and 114. You could see that the teams with practice were worlds apart from the ones without. At Silicon Valley, there were far more teams who had practice time, often from other competitions. Therefore was much more difficult.

Take 254, at Sacramento, they did not seem as tough as they were at nationals. Their drivers were new and their programs were not calibrated. In fact, their arm seemed to shaky and too quick for their arm operator. As result, their robot was capping less then our robot at times. We were happy that we were better then 254 for once. However, at silicon valley, a new end piece for their arm and a lot more practice and testing, they were far better, and easily the best robot at the Regional (kicked our butts). Still they were a an occasional quirk. Then at Nationals, from what i saw, they were flawless. My point? Teams improve over time. 254 went from good to amazing. I think, later the regionals are, the harder they get. Just going by who made it to a certain round or average score won't cut it. You have to look at the quality of the robots at the competition, a tough task indeed.

[Andy Grady]

Ahhhh good ol' doctor Joe...bringing up the stats to hoist up the midwest on the grand pedestal! Well Joe, I do believe that there are about 50 or so teams from the New England area who would take those metrics, throw them to the ground, stomp on them, and then laugh at the stats.

I do not deny that the midwest has dominated the championship tourney since about 1997...however, I think that the championships actually show about 0% indication of the actual strength of the regions or regionals in the country.

Here is my case...

There are multiple styles of play in the region. The midwest and south have always been mainly offensive regions, as where the northeast and west (growing by the year) are becomming more geared towards the defensive. The fact that New England only has one team represented in the final teams of the tournement to me is just more of an indicator of population of midwestern teams representing at the championships. More teams from offensive regions mean more of an offensively geared tourney...obviously favoring the midwest.

Its as if to say...if you were to take a team out of New England, send them to the GLR, they would not stand a very good chance based simply on style of play. I think that midwestern teams (post 1998) have shared that fate in every attempt at a championship in new england for the same reason.

I think the only true way to find out who really is the toughest region around is to duke it out on an even plane. 12 of our best versus 12 of your best...winner takes takes it all home and so on. This is what made Rumble at the Rock such a great competition in its day. At the end of that competition, there was no denying who was tops when all was said and done.

In short...you can take all the stats you want, you can take the trophies and banners and all the fun things you want to flaunt...but until our regions go toe to toe...there will be no true answer to who is the best...

Well outside of the fact that everyone should already know its the New England Region ;)

Lets find a way to do this..eh?

-Andy Grady

Disclaimer: This was meant to be a good natured post to try to stir up a good ol' rivalry between regions and maybe at some good legitimate conversation to the board. Do not take this as demeaning the midwest or any other region. In other words...relax people, this is just for fun :p

[Kate Leach]

Disclaimer: This was meant to be a good natured post to try to stir up a good ol' rivalry between regions and maybe at some good legitimate conversation to the board. Do not take this as demeaning the midwest or any other region. In other words...relax people, this is just for fun :p

Wow Andy, you relaxed them so much that they didn't even respond to your post... (or maybe they're just using the linear view instead of the threaded)

[Dave Flowerday]

The midwest and south have always been mainly offensive regions, as where the northeast and west (growing by the year) are becomming more geared towards the defensive. The fact that New England only has one team represented in the final teams of the tournement to me is just more of an indicator of population of midwestern teams representing at the championships. More teams from offensive regions mean more of an offensively geared tourney...obviously favoring the midwest.
Hmm... let's check the data. Divide the country into Midwest, South, Northeast, and West as you said:

Midwest: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and Wisconsin
Northeast: Connecticut, Delaware, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont, and West Virginia
South: Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia
West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming
International: non-US teams

Take the teams at the Championship this year, and with some Excel magic, count the number of teams in each region:

Midwest: 94
Northeast: 116
West: 44
South: 70
International: 16

Doesn't look to me like the Midwest dominates the Championships (in attendance, anyway)...

I'll leave it as an exercise to the reader to draw conclusions about what this indicates about various "regional" strategies.

.......

For those interested, I used this map (http://www.asa-cssa-sssa.org/branch/branchmap.jpg) to place the states into different regions since it seemed quite reasonable. Attached is the spreadsheet used, feel free to try it out and move states around between the list of regions if you like.

[EricH]

Let's take Andy Grady's idea of regions competing against each other. Each region (Northeast, South, Midwest, West) figures out their best robots, preferably enough for two alliances. How this is done is up to the imagination. Each region decides its alliances and sends them to a competition, say somewhere in neutral territory (a state that is not represented and is not a hot contender). At the competition, you have two divisions, and each plays round-robin style. The winners advance to a best-two-out-of-three type of playoff. The winner is declared to be from the best region. If, however, the winners are from the same region, that region is instantly acknowledged the best, and the event is over, unless someone wants to see who will win.

The winners will have the toughest regionals in their area, and it may be impossible to tell from this which regional is toughest.

[Ben Lauer]

I think the only true way to find out who really is the toughest region around is to duke it out on an even plane. 12 of our best versus 12 of your best...winner takes takes it all home and so on. This is what made Rumble at the Rock such a great competition in its day. At the end of that competition, there was no denying who was tops when all was said and done.

Andy, I think the only fair way to do it is to have 2 tournaments. And with two different numbers of teams. Don't worry, we can include the south, west, and all the other combinations of the cardinal directions.
Here are my thoughts on a "regional tournament"

Do it like congress and the senate. Have one tournament where there is equal representation of all regions. For instance, have 8 teams from each region. Have the other tournament with an equal percentage of teams from each region. For instance, if the midwest has 20% of the teams in all of FIRST, then they will have 20% of the teams in the regional.

Doesn't look to me like the Midwest dominates the Championships (in attendance, anyway)...
The only relevance I see out of this comment is that the Northeast should have more recognition than the other regions because they have more representation at the Championship.
**Note: I am a Midwesterner, and I love corn. Please don't attack me. :-)