Standings are aimed to be released by the end of every Tuesday.
With one big event hosted by Ohio, we are back after the New Year and into the second half of the College Dodgeball season.
In the 2018 Season, we have 34 technical upsets in 145 ranked matches (76.55% success rate) across 24 events. This past weekend, we had two technical upsets in fifteen ranked matches.
The Gonzalez System is a computer ranking model similar to Elo and is a rating exchange system based on research performed by World Rugby. It has been adapted by the NCDA to the demands of College Dodgeball, but can be tuned and customized endlessly to incorporate accurate data. It has been used to help determine seeds for the Nationals bracket since Nationals 2014, and was used exclusively for the Nationals 2017 bracket.
Technical Upset Spotlight
A technical upset in the Gonzalez System is when a lower rated team defeats a higher rated team. The overall success rate of the system is currently 80.65% based on 270 technical upsets in 1395 ranked matches played since 2010-09-25.
This weekend had two technical upsets over 15 ranked matches.
Ohio def UK 5-1
40.276+1 def 46.269, exchanging 1.499
With a large upset in the second round, a home court Ohio defeated a Kentucky that previously bested them 4-1 in October’s BGSU Invite. This technical upset is significant for falling in the 84th percentile. Ranking 42 of 270 technical upsets, this makes it one of the higher valued exchanges across our 1395 ranked match history.
Kent def BGSU 4-1
47.682 def 47.853, exchanging 1.017
In terms of statistically significant upsets, this one is low when viewed in comparison of other upsets. This ranks 254 of 270 technical upsets, in the 5th percentile. When viewed as a single item, this is a match between two top five rated teams with very close ratings. A one point rating exchange is very much felt by both competitors after the result, given their similar ratings. Matches like this one, where both teams are closely rated, is an important gauge on the predictive accuracy of the Gonzalez System.
Net Rating Changes
Kent was able to claim the largest net rating boost mostly for their larger value win over BGSU (an +1.017 exchange). Kent’s close win over MSU netted them a respectable chunk (+0.437). It was enough to fully negate their loss to CMU (-0.322).
Kent remains one of the few opponents that CMU can play in order to boost their rating. Even though CMU posted a 3-0 record, the current OSU and UK ratings were too low and gave the minimum rating boost of 0.010. Central is the top rated team in the NCDA and a full 2.255 points ahead of runner up Grand Valley. As the top rated team, it will always be harder to increase that gap by a sizable margin. Playing Kent and BGSU were the only statistical options for CMU, but we have to remember that tournament schedules can’t always account for even matchups. That would make for uninteresting seasons and ultimately bad for parity across the League.
After this weekend Ohio has a 12-10 record (22 games!) and their rating has fluctuated over the season, but is currently three points higher than at the start of the season. Ohio’s larger value technical upset over UK was enough to push them into a positive net rating change for their 2-2 record over the weekend. That technical upset also contributes to UK’s netting the biggest rating loss after their 1-2 record. The loss to highly rated CMU resulted the negligible minimum match value (0.010), and UK was able to get a small chunk in a win over OSU (+0.374). But being on the receiving end of a technical upset is enough to put UK’s rating lower than other teams that went 1-2 over the weekend.
SVSU posted a 3-1 record for their trip down to Athens, but their last match of the day was an statistically important one against similarly rated BGSU (47.430 def 46.732, exchanging 0.930). SVSU exchanged +.275, +.794, and +0.010 for defeating OSU, MSU, and Akron respectively. This was able to net them a small positive boost over the weekend, but not a huge amount needed to jump places in the competitive top section of the ratings.
I should point out that while Akron posted a 0-3 record, they took OSU to Overtime, making that one point away from from securing a technical upset. Akron was the lowest rated team going into this event hosting teams from all higher sections of the ratings. It sets Akron up to learn from better teams but generally not risk so much in losing to these higher ratings (-.560, -.010, -.381). Akron will be able to take this battle experience and be a much improved team for The War on their home court, I’d expect.
|↑ from 5||3||48.815||Kent|
|↓ from 3||4||48.483||JMU|
|↓ from 4||5||48.360||BGSU|
|↑ from 8||7||45.802||SVSU|
|↓ from 7||8||45.134||UK|
|↑ from 17||14||40.227||Ohio|
|↓ from 14||15||40.139†||WKU|
|↓ from 15||16||40.020*†||ZAG|
|↑ from 23||22||38.269*†||SIUE|
|↑ from 24||23||38.218||Miami|
|↓ from 22||24||38.126||OSU|
|↑ from 31||30||36.169†||NSU|
|↑ from 32||31||35.569||UVA|
|↑ from 33||32||35.442†||GSU|
|↓ from 30||33||35.364||Akron|
Movement as of 2017-11-21
* denotes a provisional rating (< 6 matches)
† denotes a team that has not played three games this season, the required minimum games needed to qualify for Nationals.
New Year, New You. Remember to stretch before you get back into the swing of your throws. Ease yourselves back into dodgeball after the winter break, so you don’t break your arms.
Your Input. Just as the system allows technical upsets to adjust ratings and improve the overall predictability, so to I welcome input in order to improve the system as a whole. What are your thoughts? Each week I try to write a bit about the system using the real examples we had occur that weekend. I also take special attention to explain concepts as if there were new readers coming into this article, or dropping explanations to previous work. Feel free to drop me a message, I’m always happy to answer any questions. See you next week.
See the Resource Center for more documentation.
Records, Master Spreadsheet: 2005-Present
Records, Individual Docs: 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018
Systems: Gonzalez Current,
Gonzalez Old, Perrone, Champ, LieblichSpec Document: Gonzalez System Spec Doc