Reply to Cap Hill - $30 daily and $70 weekend passes?

Forum Index » Volleyball

Reply to "Cap Hill - $30 daily and $70 weekend passes?"

Subject:

Emoticons





More smilies

Text Color: Font: Close Marks

[quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous]That's why I don't understand why every year there are rumblings about conspiracies to rig the tournament in favor of Metro. The Metro Travel teams should always be in contention to win most of the open divisions. The Metro Travel teams should be seeded highly because they are amongst the best teams inthe tournament - that's how seeding works. 
[/quote]

The issues with seeding aren't about where Metro is seeded [b]its how the other teams are seeded relative to Metro[/b]. In a typical 48 team tournament the pools are set up so that the first pool should be the #1, #24, #25 & #48 teams in the tournament.  The second pool is #2, #23, #26 and #47, etc.  This is universally the approach in every tournament except power pool tournaments (like NIT).  After day 1 the top finisher in pool 1 plays the 2nd finisher in pool 2 (which should be #1 vs #23) while the #1 finisher in pool 2 plays the #2 finisher in pool #1.  Same for pools 3 & 4, etc.

Importantly, teams in pools 3 & 4 can't compete against pool 1 & 2 until day 3 in gold bracket.  So if you want to make sure you get to gold bracket and get a good seed, you don't want to play any of the high ranked teams on day 1 or day 2.  Being ranked #1 in the tournament takes care of that on day 1.  But what about day 2?  Sometimes good teams have a "off" match and end up coming out 2nd in a pool that they should have won.

If you had bias in seeding you would try to make sure than none of the top teams could play you on day 2, regardless of how their day 1 went. And you would try to seed the tournament so that the best teams compete against each other and knock each other out, so you don't have to face them later.

Here are the relative ranks based on current AES ranks of the teams in the AES ranks of the teams in the 15 Open division in pools 1-4:

Pool 1: Metro Travel: #1 (AES #3), American 15 Red: #31 (AES #1134)
Pool 2: 540 VB 15 Elite: #20 (542), VA Elite 15: #12 (283)
Pool 3: Pittsburgh 15 Elite: #25 (747), Loudoun Elite 15 Tony: #9 (235)
Pool 4: MDJRS 15 Elite Black: #2 (57), VEVA Fury: #16 (449)

Notice how Pool #4 has the #2 overall team while Pool has 2 no teams in the top 10?   In fact, the #2 seed in the overall tournament (540 VB) isn't even ranked in the top 500 nationally (and lost head to head matches against Blue Ridge, Yorktowne, and VEVA Fury -- all of whom are seeded below them in the tournament).  And they already played Metro at Charm City and lost in a blowout.  Meanwhile, pool 4 has the second overall seed ensuring no matter what happens Metro travel is guaranteed not to play them on day 2.

Also, if the first four pools only have 2 of the top 8 teams (when it should be 4/8), that means the remaining pools must have more top teams competing against each other (6/8 in this case meaning that those teams), effectively "stacking" the competition on the other side of the bracket from Metro.

Similar seeding changes where made in 16 Open (just showing AES ranks this time)
Pool 1: Metro Travel (9), Chicago Elite 16 (1018)
Pool 2: Ultimate VBC 16 Gold (840), ECJVC 16 National (775)
Pool 3: Blue Ridge 16 Blue (73), CHAVC 16 Black (582)
Pool 4: MVSA 16 Sparks (170) , CALI 16 Black (NR-1173 last year)

Pool 2 doesn't have a highly ranked team in it.  The #2 seed (Ultimate VBTC) has 15+ teams ranked above them in AES in the tournament. The second highest ranked team in the tournament per AES (Blue Ridge) is in pool 3, guaranteed not to play Metro on day 2 or until the semifinals on day 3. There's also the same issue with stacking teams in the lower pools.

17 Open 
Pool 1: Metro Travel (19), Rival 17 Black (384)
Pool 2: Chicago Elite 17 (326), Huskies 17U Premier (NR-772 last year)
Pool 3: 757 17 Black (789), VA Juniors 17 (64)
Pool 4: TVC 17 Black (48) , 575 17WSE Taylor (NR)

Again, the #2 (TVC) and #3 (VA Juniors) teams overall are in pool 3 & 4, while pool 2 is the weakest of the 4, and the same issue with team stacking.

There are other inconsistencies in the other divisions.  Its true seeding is an inexact science. AES ranks aren't perfect (although they do have a strong correlation to future performance against teams that are significantly higher or lower than your ranking).  Tournament directors can and should move teams up or down based on additional info not available in AES.  But systemic inconsistencies like the 3 above are statistically unlikely, supporting a basis for the claims that Metro gets beneficial seeding at Capitol Hill.  It doesn't happen in every division every year but it does happen frequently enough it appears there is some bias, either intentionally or unintentionally.[/quote]

You have too much time on your hands, I fear. [/quote]

I don't know how to read this message. Are you trying to be dismissive or to ridicule the PP who did the research? You have better things to do with your time or your time is too important? Nobody asked you to waste your time digging up data - maybe you wouldn't even know what to do in the first place. Just because you don't have the time to look up pools or you don't want to do the work, it doesn't mean that others feel the same way about this issue. Instead of trying to ridicule the PP's work, you could have said "Thank you for doing this research, this is eye-opening."

I have to admit that I kept dismissing the idea that the tournaments are rigged to favor Metro. This PP completely changed my perspective and now I understand that - even if you are #1 in the region - you can screw up from time to time. I also understand how the tournaments can be rigged to favor the best team in the region. And I believe that PP provided enough evidence to demonstrate that this is not by chance, but on purpose. Thank you, PP - this was eye-opening. [/quote]
Inherent in all this speculation is that the results of every match can be reliably predicted using unreliable data sources. While I am not of the belief that the AES ranking system has no value (there are many knowledgable people who feel this way), I also don't believe it's accurate enough for anything other than really broad comparisons.  A team ranked in the top 100 is probably better than a team ranked 1,000, but I don't think it's reliable enough to say with any confidence that a team ranked 50th is clearly better than a team ranked 75th or even 200th depending on the time of the season. And even if it were reliable, does that mean the outcome of any match is predetermined with the higher ranked team winning every time?  Obviously not.

I watched some of Metro's 17 and 18 Open gold bracket matches at Cap Hill yesterday and from what I saw, it didn't seem like the best two teams made the finals in both divisions. From what I observed, Northeast 17.1 and Northeast 18.1 seemed like the best teams other than Metro 17 and 18 Travel and they came in not seeded very high and also not ranked very high in AES.

For 17s, Northeast 17.1 was initially seeded 9th in 17 Open but won all their matches on Saturday and Sunday until meeting Metro in the 17 Open quarterfinals, where they lost in 3 in a very good match.  Other highly placed teams in 17 Open included NYC VBA 17 Westside who lost to Metro in 3 in the semifinals, and Jupiter Elite 17E who lost in 2 to Metro in the finals.  Here's how they are ranked in AES as of today, their initial CHC seeding, and their finish place:

Metro 17 Travel, 27 in AES, seeded 1 at CHC finished 1st
Northeast 17.1, 863 in AES, seeded 9 at CHC, finished tied for 5th
Jupiter Elite 17E, 459 in AES, seeded 8 at CHC, finished 2nd
NYC VBA 17 Westside, unranked in AES, seeded 5 at CHC, finished tied for 3rd

For 18s, the story is similar. Northeast 18.1 was initially seeded 12th in 18 Open but won all their matches on Saturday and Sunday, including beating Metro 18 Travel in Sunday pool play.  This resulted in them meeting Metro in the 18 Open semifinals, where they lost in 3 in a back and forth match. VA Juniors 18s made it to the gold final, which Metro 18 Travel ultimately won in 2, but the first set was close with VAJRs leading for almost the entire set and Metro making a late comback. Here's how these teams are ranked in AES as of today, their initial CHC and their finish place:

Metro 18 Travel, 156 in AES, seeded 1 at CHC finished 1st
Northeast 18.1, 391 in AES, seeded 12 at CHC, finished tied for 3rd
VA Juniors, 159 in AES, seeded 9 at CHC, finished 2nd

Obviously watching a team in a single match does not give a complete picture of how good a team is in the overall national landscape, but from what I saw the Northeast 17s and 18s teams were among the top handful of teams there and their record over the course of CHC seems to validate that.  But their rankings in AES don't reflect that and their seeding to start CHC doesn't seem too out of line given the AES rankings. It does seem like maybe VA JRs should have been seeded higher if AES ranking were the sole factor in CHC seeding, but that said VAJRs got lucky and had a much easier path to the finals in the gold bracket, getting a first round bye and then most of the other best teams being on the other side of the bracket.

My point is that none of this is an exact science and that in most cases, the best teams will still rise to the top. Particularly at a tournament like CHC which is not super competitive and where in most of the Open divisions there are a handful of teams that are far better than the rest of the field.  The reality is at CHC, the top teams should coast through day 1, day 2 should be a little tougher, and day 3 begins the real tournament in bracket play. 
[/quote]

Thanks for the info here, really appreciate the thought and effort put into this.
Regarding AES predictive power, statistically the bands of accuracy for AES rankings are historically in the following ranges: 1-50, 50-150, 150-300, 300-600, 600-1000, 1000-2000 and 2000+ (for the U14-U17 ages, they move for the smaller age groups).  There is a strong correlation when you play someone outside your band.  Obviously the accuracy increases as more games are played--especially when those games cross regions--but even early in the season teams there is a correlation.  Haven't run the data recently but historically the top 50 wins 70%+ of the matches against the next band and 80%+ when you get more than 2 bands out. That pattern holds all the way through the top 300-600 or so and then becomes very predictive (90%) when you get large differences between teams (e.g. 100 vs 1000).

On Cap Hill, you are correct that the top teams should coast through day 1 and day 2, with day 3 becoming much harder.  Capitol Hill is a big outlier in this though especially in the pools seeded in the crossovers against Metro.  The top teams generally don't coast through the first two days.  Here are the results from the top 2 seeds in pool #2 in each division, linking to the earlier post about seeding bias.  Remember that this is the pool that crosses over to Metros pool on day 2 because that is the pool that seems to be seeded in favor of giving Metro an easier route to the gold bracket.

15 Open:
#1 seed pool 2, 540 VB 15 Elite: 3-0 day 1, 1-2 day 2, 0-1 day 3.  Overall record 4-3, placed 15th
#2 seed pool 2, VA Elite 5: 2-1 day 1, 0-3 day 2, 0-1 day 3.  Overall record 2-5, placed 23rd.

16 Open:
#1 seed pool 2, Ultimate VBC 16 Gold: 2-1 day 1, 1-2 day 2, 2-0 day 3. Overall record 5-3, placed 17th
#2 seed pool 2, Columbia 16 Black: 2-1 day 1, 0-3 day 2, 0-1 day 3. Overall record 2-5, placed 29th

For these two brackets, the top 2 teams in pool 2 went a combined 2-8 on day 2 and none of them reached the gold bracket playoffs. These also happen to be the pools with the largest statistical variance between their AES ranks and their seeding in the tournament. The average AES rank of the four teams was 484.

For 17s, the top 2 pool 2 seeds did get to gold, but lost in the first and second round and ended with a combined record of 1-2.

For 18s, only 1 pool 2 team (540) got to the gold bracket. They went 1-1 on day 3.

[/quote]

Options

	Disable HTML in this message
	Disable BB Code in this message
	Disable smilies in this message

Review message

Search Recent Topics Hottest Topics