Posted by Joe Johnson, Engineer on team #47, Chief Delphi, from Pontiac Central High School and Delphi Automotive Systems.
Posted on 3/14/99 7:43 PM MST
Okay, here is the deal:
This is a long message, but I hope you will stick with me on this one.
Motorola had 10 qualifying rounds: great
California had 12 (I think) qualifying rounds: even better
Florida had 12 qualifying round: even better two
Philly had 6 qualifying rounds: not even close to enough
I don’t know how many will be at the New England and Mid-Atlantic, but I guess that it will be 10 or 12.
Great Lakes will likely have 6 : Again, not even close to enough
Now, on to the Nationals:
Usually reliable sources claim that the current plan is to have only 4.
I ask you, 4?
They may as well just throw darts to pick the top 8.
I am serious, very serious. Think about it. A top 20th percentile team (i.e. a very good team) has a 1 in 5 chance each match of being partnered with a bottom 20th percentile team (i.e, a not very good team). In this case, the good team is essentially playing 2 on 1. So far so good.
Now, what chance is there that a top 20th percentile team will be paired with nothing but bottom 20th percentile teams for all their qualifying matches? It is approximately (1/5)^4 or 1 in 525.
This doesn’t sound too terrible until you realize that there are 40 top 20th percentile teams. Essentially, one time in 13, the situation will occur. Not too bad of odds, you say, it will only happen on average once in every 13 FIRST Nationals, and that one time, the team will most likely be picked up by one of the top 8.
Yes and no. In a field of 200 teams, it is going to be very hard to stand out of the crowd, especially if you lost every match and are very poorly ranked.
The situation gets worse when you realize that only 3 such nightmare matches are likely enough to spoil any chance such a team has at making it into the top 8.
The chances of this are considerably more discouraging: 4 X (1/5)^3 or about 1 chance in 30.
With there being 40 top 20th percentile teams in a 200 team tourney, FIRST is virtually assuring that the randomness of the seeding will keep one or two team that might otherwise have a legitimate shot at the top 8.
The story gets worse when you ask yourself how many middle of the road teams (say 20th percentile to 80th percentile) have great tournaments simply because they end up with three or more top 20th pecentile teams for partners.
Again, the odds are 1 in 30, but in a 200 team tourney, there are 120 teams in the middle! At the Nationals, 4 of these middle teams have a legitimate chance to make it to into the final 8 just because they were lucky enough to have 3 more great partners!
The only solution is to have more qualifying rounds.
I feel that this is a serious threat to the integrity of the Nationals.
I hope that there is time to change the plan for the Nationals.
I urge you to “write your congressperson.”
Joe J.
P.S. Note that the proposed plan is a step back even from prior Nationals. They are proposing FEWER matches at the Nationals than they had last year. Not only fewer for all teams (all teams had at LEAST 6 matches in prior years – 4 seeding round and a minumum of 2 in the tourney) but fewer overall matches! Note that while there are 33% more teams at the Nationals this year (~200 vs. ~150), but this year’s format lets 33% more teams play each round. Therefore, by having most teams only have 4 matches at the Nationals rather than 6, FIRST is actually proposing a reduction in the absolute number of matches of over 33%. This is progress?