Seeding Calculations

[archiver]

Posted by Raul.

Engineer on team #111, Wildstang, from Rolling Meadows & Wheeling HS and Motorola.

Posted on 9/7/99 3:07 PM MST


In Reply to: Re: If you think that was complicated. try this... posted by P.J. Baker on 9/7/99 5:50 AM MST:



: Raul,

:
: With a 6 match schedule, I can't quite agree that won lost record should be the most important factor in seeding. I do think though, that this year it was not important enough. It probably was possible for a 2-6 team to seed ahead of a 6-0 team. That's not right.

: I'm perfectly comfortable with a 5-1 team, or even maybe a 4-2 team seeding ahead of a 6-0 team. A team that should have been 4-2 is just as likely to get 2 lucky wins and go to 6-0 as a team that should have been 6-0 is to get 2 unlucky losses and slip to 4-2.

: This is why I think that it is important to also look at the scoring and defensive capabilities of the team. If we played lots more matches, I'd be with you 100%. With only 6 though, I think that the seeding should be based on as many measureable factors as possible (won/loss, ave. offense, ave. defense). What do you think?

: P.J.

I agree and therefore support a system as described in a recent posting by Mike Aubry.

Raul

[archiver]

Posted by Austin Martus.

Other on team #47 from son of pchs coach.

Posted on 9/7/99 5:56 PM MST


In Reply to: Re: If you think that was complicated. try this... posted by P.J. Baker on 9/7/99 5:50 AM MST:



why dont they just establish a win loss record from the first regional you go to and keep it with you to the very end and for those who could not attend as many reagional like give them like and even record like if the most wins you could have had if you went to most regional possible for example 10 and you would get a 5-5 record type thing mabey like either added to you established record if you only went to one regional or all of the example 10 possible if you didnt go to any regionals

this would give you a bigger base but i dont know how effective it would be

just a thought fragment thing needs work


austin

[archiver]

Posted by Joe Johnson.   [PICTURE: SAME | NEW | HELP]


Engineer on team #47, Chief Delphi, from Pontiac Central High School and Delphi Automotive Systems.

Posted on 9/6/99 7:04 PM MST


In Reply to: Re: not bad... not bad... posted by Bethany Dunning on 9/5/99 9:38 PM MST:



While the games of golf and tennis are fairly easy to explain, the ranking systems in both sports are VERY complicated.

Even college football is getting a 'rocket science' ranking systems.

I think that the ranking system can be as complicated as we want to make it be. As long as we can understanding it amongst ourselves. Even if we all don't fully comprehend all aspects, all that is really required is the 10 cent version (e.g. Scoring high is good, keeping your opponent score low is good, etc.)

Joe J.

[archiver]

Posted by Tom Vanderslice.

Student on team #275, ORHS/AST/Hitachi, from Academy of Science and Technology and Hitachi.

Posted on 9/6/99 7:51 PM MST


In Reply to: try explaining rankings in tennis or golf... posted by Joe Johnson on 9/6/99 7:04 PM MST:



nascar has a wierd ranking system based on a very straightforward concept...
the point of every race is to be ahead at the end...

but the ranking system is based on how many laps you were in the lead and how many laps you actually completed
and all this other stuff...
(maybe we could look at the splits like every 30 seconds who's in the lead an you get bonus for taht....

tom

[archiver]

Posted by Dan.

Student on team #10, BSM, from Benilde-St. Margaret's and Banner Engineering.

Posted on 9/6/99 9:25 PM MST


In Reply to: try explaining rankings in tennis or golf... posted by Joe Johnson on 9/6/99 7:04 PM MST:



I'm afraid complicated rankings are required, and we should all be grateful for that too.
The fact is that FIRST is trying to make this as fair as possible while using a very small chunk of results. This isn't baseball where we play a hundred plus games a year, we are competing for a weekend and every effort to fairly rank us based on those results must be made. And for that reason FIRST must find and use as much information as possible. If they can step beyond 'wins' and 'losses' then they can only make things more fair.
:-Dan

[archiver]

Posted by Tom Vanderslice.

Student on team #275, ORHS/AST/Hitachi, from Academy of Science and Technology and Hitachi.

Posted on 9/2/99 7:37 PM MST


In Reply to: That doesn't sound normal posted by P.J. Baker on 9/1/99 9:38 PM MST:



here we go...

we take the distribution of scores...
we force fit the distribution to a bell curve... (not sure exactly how...bu ti know there's a way to do it)...
then you score qp's based on this 'normalized score'...
your person on your team in charge of watching the score board goes insane trying to figure out what the heck is happening with scores...


Tom

[archiver]

Posted by P.J. Baker.

Engineer on team #177, Bobcat Robotics, from South Windsor High School and International Fuel Cells.

Posted on 8/31/99 6:21 AM MST


In Reply to: not a fair comparison... posted by Joe Johnson on 8/30/99 5:16 PM MST:



Joe,

I know that you guys at least have all of the match scores from Great Lakes. Why not try it out on those scores?

P.J.

[archiver]

Posted by Jason Morrella.

Coach on team #254, Cheesy Poofs/Bay Bombers, from Broadway High and NASA Ames.

Posted on 8/30/99 2:20 AM MST


In Reply to: Seeding Calculations posted by Joe Johnson on 8/28/99 8:23 PM MST:



Joe,

I agree 100%. I think taking the highest & lowest score out of the average would provide a much more accurate seeding. My only condition would be that they find a way to go back to every team getting a minimum of 8 matches instead of 6. If each team could get 8 matches, I feel that averaging the middle 6 would provide a fairly accurate seeding.

About the final multiplier, I must recommend the system I saw used at the Anaheim Rumble held in June. Instead of the winning alliance getting a triple multiplier, each team on the winning alliance got a 300 point bonus added to their score. This made the scores much more accurate in terms of ability because it rewarded teams which consistantly won due to great defense & strategy the same as it rewarded teams which just went for high scores with no concerns for defense in qualifying rounds. I was somewhat skeptical when first told of the scoring system, but was a firm supporter of it after watching a few matches. I would highly recommend something similar if alliances are used again next year.

Hope everyone is looking forward to the start of school as much as I am!!! (really, no sarcasm there what-so-ever) Just means we're getting closer to the 2000 competition and the new & improved Chessy Poofs!

Jason

[archiver]

Posted by Jon.

Engineer on team #190, Gompei, from Mass Academy of Math and Science and Worcester Polytechnic Institute.

Posted on 8/30/99 2:53 PM MST


In Reply to: Re: Seeding Calculations posted by Jason Morrella on 8/30/99 2:20 AM MST:



this is interesting... a simple bonus pointage system instead of a multiplier... small (points) victories are rewarded instead of super pointage... more like the NBA/NFL/NHL/et al

[archiver]

Posted by Jason Morrella.

Coach on team #254, Cheesy Poofs/Bay Bombers, from Broadway High and NASA Ames.

Posted on 9/2/99 1:00 PM MST


In Reply to: Re: Seeding Calculations posted by Jon on 8/30/99 2:53 PM MST:



: this is interesting... a simple bonus pointage system instead of a multiplier... small (points) victories are rewarded instead of super pointage... more like the NBA/NFL/NHL/et al

To clarify: the multipliers (2x, 3x, 3x) were still in effect for raised floppies, puck location, and puck control. Team could still max out with 540 points. The only difference was the final victory reward.
It evenly rewarded teams for winning with 300 bonus points instead of a 3x multiplier. A team which won with a total of 400 points would finish with 700. A team which won with a total of 70 points would finish with 370.
So in the qualifying seedings, two winning alliances with scores of 400 and 70 respectively would have 700 and 370 with this system, not 1200 and 210 with the 3x multiplier system.
I think most agree that the point is to have a good strategy, execute it, and win. It seems that in the above example, the winning teams being seperated by 1000 points is basically punishing the team which won with a defensive strategy.
It worked well at the exhibition in Anaheim and I thought it was worth throwing out for people to consider. So consider away.

Jason

[archiver]

Posted by mike aubry.   [PICTURE: SAME | NEW | HELP]


Engineer on team #47, Chiefs, from Pontiac Central.

Posted on 9/5/99 2:06 PM MST


In Reply to: Seeding Calculations posted by Joe Johnson on 8/28/99 8:23 PM MST:



Joe and many others -
Sorry I joined the conversation late, but I have been thinking that maybe seeding
could be revised such that scoring (offensive points scored) would only be one factor
that would contribute to the seeding calculation. As I see it, scoring should be very,
very, easy for everyone to figure out as the game progresses along toward the final
seconds. I think that is the only way the spectators that are not as involved as some
of us lunatics will ever be able to understand the score. Every game that I have ever
seen relied on a relatively simple scoring system. Examples include: Baseball (1 score
for every person crossing home plate) Soccer and Ice Hockey(1 score for every ball or puck that
goes into the goal) Football (Touchdown equals 6 pts, extra point 1 or 2 depending on
if kicked or carried over the goal line, 3 pts for a field goal) Basketball (2 or 3 points when
the ball goes thru the hoop, 1 point for each free throw) Enough of that - you guys already
understand. Now, this is where it gets tricky! Winning a game is recorded as a win, loss, or
tie. Rankings are developed using formulas that include many variables. (Ignoring the most
obvious type - personal favorite) Seeding is just like rankings. Seeding can be a combination
of factors that include things such as: average offensive score, average defensive score, points
given for beating a higher ranked team, points taken away by losing to a lower ranked team, etc.
But whatever the seeding or ranking is determined by - as long as everyone knows the formula
at the start of the season I'm sure we will all be able to adapt. The conversation can still be
stimulating discussing the possibilities though! I do not understand the necessity to create
a scoring system that requires a calculator, mathematician, and statistician to be able to determine
the outcome of a game. You can have the multipliers for seeding factors, I say that if you want to get
the common persons attention - Keep the scoring simple ! That has always been my motto!
Seeding on the other hand can be developed to encourage or discourage offense or defense, as
well as, winning and not winning. Make as complicated algorithm as needed to - but I think the
2 issues need to addressed independently. I would develop tha seeding algorithm simularly to either
PJ's or Joe J's that were shared elsewhere. I think the issue is one of gaining a better method that
we feel will lead to a more accurate ranking, and that all depends on what one wants to weigh heavier
in the mathematical equation. That's my thoughts, I welcome others!
Thanks, Mike

[archiver]

Posted by Bethany Dunning.   [PICTURE: SAME | NEW | HELP]


Coach on team #163, Quantum Mechanics, from International Academy and Quantum Consultants/EATON/ITT Industries.

Posted on 9/5/99 9:27 PM MST


In Reply to: What if seeding wasn't solely dependent on scoring? posted by mike aubry on 9/5/99 2:06 PM MST:



But as long as you are using sports as an analogy, let's take it this direction: when it comes time for seeding
for playoffs in basketball, hockey, etc, it is your win/loss average that gets you where you want to go. Take
the NHL (my team is snickering right now - they know my love for hockey). Wins are worth 2 points, ties are
worth one, losses worth nothing. The team with the most points is the top seed. No opinion, no how many
goals did you score, no penalty minutes. Just whether you won or lost.

[archiver]

Posted by Tom Vanderslice.

Student on team #275, ORHS/AST/Hitachi, from Academy of Science and Technology and Hitachi.

Posted on 9/5/99 10:42 PM MST


In Reply to: Re: What if seeding wasn't solely dependent on scoring? posted by Bethany Dunning on 9/5/99 9:27 PM MST:



as good as that sounds...in hockey you are dealing with what? 20 teams...(i'm no hockey expert...i'm from texas...so bear with me...)
now...that's ok at regionals...now..fast forward to nationals...you have 300 teams...now...how many of htem are 6-0?...5-1?
how do you seed them??

tom

[archiver]

Posted by Raul.

Engineer on team #111, Wildstang, from Rolling Meadows & Wheeling HS and Motorola.

Posted on 9/5/99 11:02 PM MST


In Reply to: Re: What if seeding wasn't solely dependent on scoring? posted by Tom Vanderslice on 9/5/99 10:42 PM MST:



: as good as that sounds...in hockey you are dealing with what? 20 teams...(i'm no hockey expert...i'm from texas...so bear with me...)
: now...that's ok at regionals...now..fast forward to nationals...you have 300 teams...now...how many of htem are 6-0?...5-1?
: how do you seed them??

: tom

Use points for the tie-breaker. You could use the same system for QP's, but make won-loss record the first criteria.

[archiver]

Posted by Bethany Dunning.   [PICTURE: SAME | NEW | HELP]


Coach on team #163, working on a new name!, from International Academy and Ford Motor Company.

Posted on 9/6/99 12:07 PM MST


In Reply to: Re: What if seeding wasn't solely dependent on scoring? posted by Tom Vanderslice on 9/5/99 10:42 PM MST:



I'm not saying use the system they use in hockey, basketball, etc. But there has to be something simpler than some of the ideas posted here.
And Tom, you do realise that there is hockey in Texas? In fact, y'all have a Stanley Cup (as much as us Canadians sometimes resent that fact! :-)