Seven On: Nationals Seedings, URPI

by | October 10, 2013, 10:39am 0

If you somehow missed that painful hour of USAU’s Triple Crown Tour Twitter slowly releasing the seeding, they’re thankfully now out in full for all of the divisions. I put together a handy spreadsheet with previous pool matchups for the discussion below. So be sure to check it out. The spreedsheet includes Women’s and Mixed, though I’m going to be talking about Men’s.

1. Pool A

San Francisco Revolver, NY PoNY, DC Truck Stop, Florida United

Revolver takes the one seed of the tournament, after a regular season which saw them lose only four games, and take the top spot in the USAU rankings. There was thought that maybe GOAT or Doublewide could sweep in and take the top spot from Revolver, but the U.S. Open winner was able to lay it down at Southwest Regionals and hold on to a win.

Revolver’s pool doesn’t have much competition either, with PoNY coming in as the 8th overall seed and 2nd in the pool. None of the competition is familiar to Revolver this season either, they didn’t play a single game against their pool play opponents this season – the only team at the top of a pool that can say that.

PoNY on the other hand has a pool they can’t be looking forward too. The two remaining teams in this pool are teams that gave NY trouble during the summer. Truck Stop was the main culprit, beating the NY boys twice this season (and also giving them trouble last season as well). But based on their many victories, I don’t think this year’s Nationals will be a repeat of last year, where PoNY didn’t win a single game. An upset between those lower three teams is probably the best action we’ll get out of this pool – Florida is hoping it can show that the newcomer to the scene can compete with the two teams that are among the top consistently.

2. Pool B

Toronto GOAT, Seattle Sockeye, Minneapolis Sub Zero, Columbus Madcow

GOAT fell from what some had as a potential #1 overall seed to the 2nd seed, and with that came a harder pool. At least depending on which of the lower teams show up. Sockeye sits as the 7th seed overall, and faces a similar situation. The two bottom teams, Sub Zero and Madcow, are virtually unknown to the top seeds – there was just one game played against GOAT, a 13-12 victory for GOAT over Sub Zero at the Chesapeake Invite. That was the tournament Sub Zero shocked everyone by winning. Madcow, on the other hand, didn’t win a game, but still looked dangerous. Columbus had close games with practically every opponent. It might not warrant the ‘pool of death’ moniker, but Pool B could go any way should the right Sub Zero or Madcow team show up to play. If not, look for the GOAT vs. Sockeye game on Thursday night to a big one, as the 2nd place finisher in the pool will have a much less favorable path through the Championship bracket.

3. Pool C

Texas Doublewide, Boston Ironside, NC Ring of Fire, Santa Barbara Condors

Three of this pools’ teams were in the 2012 semifinals (sans Condors), with Doublewide advancing and winning last year’s championship. All four teams have a strong history, though the Condors are hoping to make a new one – and boy were they given a great opportunity to do so. This might just be the 2013 pool of death. Doublewide vs. Ironside will be built as a game to watch, but I think you have to pay attention to what Ring of Fire does. As has been talked about, how will teams approach the pool play games? If it all goes to seed, Ring does not have a favorable matchup in pre-quarters, so I think they’ll be trying their hardest to finish within the top two. But obviously they have two teams that are pretty good sitting above them.

Doublewide and Ironside have played each other twice this season, splitting the season series with a win apiece – one a 10-15 victory for Ironside, the other a 15-9 victory for Doublewide. The pool winner gets the most favorable matchup going into bracket play, with the fourth place team of the B pool as their opponent.

The Condors don’t have any games against these three teams from the regular season, and should be looking to catch a team off-guard (perhaps Ring at the end of the day).

4. Pool D

Denver Johnny Bravo, Chicago Machine, Atlanta Chain Lightning, Vancouver Furious George

If you thought the C pool was bad, take a look at this one. Bravo and Machine as the #4 and #5 seeds overall, and Chain Lightning sitting quietly at #9 overall. Oh and who can forget unknown Furious George as the last seed in the tournament (though I think many would agree, the Vancouver boys won’t find themselves there when everything is said and done).

Furious is the unknown in this pool. There are no previous matchups between Pool D teams and Furious either. It may not have even mattered as they look to be a much better team than the shortened roster we saw at Terminus months before. That makes getting a good read on how they finish in this pool perhaps the most difficult.

Bravo, Machine and Chain have all seen each other this year, and there is no clear consensus on who wins the seasonal matchups. Bravo beat Machine, but lost to Chain, and Chain and Machine split their season matchups. It will be quite the wild ride for sure. Winner of the pool has the most favorable route in the championship bracket, though (potentially) facing one of Doublewide, Ironside or Ring in quarterfinals isn’t that appealing. Second and third may like their odds in the pre-quarters, but quarterfinals will not be a walk in the park either. Whoever finishes last can probably be counted out against what will most likely be Revolver. Look for a mad rush to not finish in that spot.

5. The Trouble in Predicting

In talking with Skyd’s Zack Smith, he pointed out a difficult part in making predictions – something Skyd and Ultiworld are hard at work doing in preparation for the USAUltimate Pick’Em contest. Pool play is going to be approached differently by every single team. Sockeye may see their game against GOAT as a must win for an easier route to bracket play, or they may not. The same could be said of the Ghosts vs. Slow White matchup on the Mixed side, or Heist vs. Riot matchup on the Women’s side. “I think if everyone was vying for the top spot in the pool, it would be easy to give our opinion,” said Smith, continuing “but how can you guess who will use the first half of a game as a feeler, or who will tank from the start…will anyone tank?”

Another reason to be worrisome about predicting – we still don’t know how seriously teams will take pool play. As pointed out, there are some big matchups that will determine where teams sit in pool play. If you look again at the competition schedule USAU released, for all divisions, the end of regular season rankings are going to mean quite a lot if your team is knocked out before semi-finals. “5 through 8 are quarterfinals losers in order of final regular season ranking. 9 through 16 are pre-quarters losers in order of final regular season ranking.”

Woah. That is big news, it weighs the regular season rankings a lot. If Revolver and Furious hold seed to prequarters, and Revolver moves on, Furious would be facing a finish no higher than 13th place. Or if there is an upset in prequarters, say #3 seed Doublewide, they could find themselves no higher than 7th and barely making the Pro Flight for next season. That’s a big change, and one that I’m under the impression teams did not know about going into the regular season. I don’t think it would’ve changed too much of what we saw at tournaments in terms of wins/losses, I just don’t see teams trying harder or lesser to play with the regular season rankings that much, but still something that would’ve been nice to know.

Speaking of that though, Sludge Online has a couple of cool charts comparing how each team finished in the ranking at the end of the 2012 season, and where they find themselves seeded this season. Overall across the divisions, there are 10 new teams to the top 16, PoNY and Bravo had the biggest jumps at +8 from the year before, and Ring of Fire and Odyssée had the biggest fall at -7.

6. What to Know About the (Bracket) Tournament?

As discussed ad nauseam in past Seven On’s, with the new non-power pools format, I’ve been thinking what other stats we would want to see as fans and ‘experts’, going into the big weekend. Typically on ESPN before the NCAAB tournaments you see how much a team has scored against opponents, what their rating percentage index is (RPI), and a number like player stats. If you look in the spreadsheet I linked earlier, I attempted to go through and pick out wins and losses for each Open team that ‘mean’ something – either in how the teams USAU ranking or seeding ended up – you can find that in the spreadsheet titled ‘Open 2’. It’s a largely useless thing overall, since it came down to opinion on which game for some. But it is all interesting nonetheless.

RPI could be interesting to figure out though, it might help us see the tournament could be seeded in a different way, or if we can expect any upsets to happen. For those unfamiliar, RPI  is used to rank teams based on their wins, losses and strength of schedule. Where it gets tricky though is how to calculate it and have the numbers produce something tangible in ultimate. The formula that the NCAA uses for most of its tournaments that consider RPI is the following: RPI=(WP x .25)+(OWP x .50)+(OOWP x .25). WP stands for Winning Percentage, OWP for Opponents Winning Percentage, and OOWP for Opponents of Opponents Winning Percentage. The NCAA for basketball also goes in and weights game differently depending on if they were won at home or away, etc – check out the Wikipedia page for a full explanation of it.

If we are going to take RPI and make it into something useful here for ultimate, I don’t think we can go into weighting everything like home and away wins and losses – mostly because those aren’t really a factor. It also gets really tricky when considering opponent’s win percentages, because of all those times when teams didn’t face teams within the top 16, or even top 30 I’d say – wins that could artificially inflate their numbers, or at least in theory could. Take the bottom four teams on the Open side for example. Those four teams have a combined one previous game against a pool play opponent, and 16 combined games against teams that made it to Nationals. Compare that to Revolver who has 14 games against teams that made it to Nationals – I think it’s safe to say their results could throw off a calculation of RPI that is done in the traditional NCAA format. Now is probably a good time to say that I haven’t done much math since high school, I was a history major in undergrad, and statistics wise I took AP Stats in high school, but that was some years ago and I may or may not have passed the AP exam at the end. So if any of my math doesn’t check out, or make sense below, feel free to contact me and I can work on adjusting it to make sense.

Here’s what I’m thinking, to calculate Ultimate Rating Percentage Index or URPI: URPI = (WP x .3) + (OWP x .4) + (OOWP x .3). But a key part of this is that we are only taking into account each team’s winning percentage when facing other teams that qualified for Nationals. It’s important to make a distinction as including all 114 teams that were in the final USAU rankings would not only be a pain to calculate, but could throw off our results skewed towards teams that had good seasons against bad teams. In order to make sure it isn’t skewed towards those teams that did play a lot of teams within the 16 teams left, I weighted the OOWP the same as a team’s WP, and their OWP a bit more. If you feel like it should be weighted differently, the basic raw numbers are in the spreadsheet under ‘Open 3’, feel free to manipulate and tell me where I screwed up in any of my calculations. Really, I welcome it.

So to wrap it all up with some actual numbers: my calculations there tell us Revolver had the strongest URPI and Madcow had the weakest. It’s worthwhile to point out as well that Johnny Bravo had the next highest URPI, while both Ironside and PoNY fall a bit from their seeding, while the Condors, Sub Zero and Chain would also rise up. This could be useful in making predictions too; when taking into account past matchups, and this URPI number, we should get a clearer picture of how each team is going to finish. Key word is should though, it would still just be a prediction of course – what Zack Smith brought up in #5 still holds a lot of weight, we don’t know how these teams are approaching the new format. Next season, it could prove to be more useful though as we’ll have an idea of how teams will approach it.

And once again: this is nowhere near perfect. I know that. The statistic itself, let alone the formula I used, are all flawed with the small sample size for sure, and aren’t the only judge of a team. Just wanted to try and give us something new to talk about with the new format.

7. Women’s and Mixed Division

As said, if you click through to the spreadsheet linked at the beginning (here it is again, ya bum), there are the previous pool matchups and their results put out in a somewhat easier to read form than score reporter. I’d like to go through and do a spreadsheet similar to ‘Open 2’ and ‘Open 3’, but between time and not knowing each of the divisions well enough for a key matchup’s spreadsheet, I haven’t yet been able to put one together. If anyone is feeling adventurous and would like to try and tackle that, go for it – the formula is above. It just takes a bit to sit through and get all the numbers; a lot of switching between the tabs. Ultiworld had a rapid reaction piece to both of those divisions, you can click through here for the Women’s and Mixed divisions.

After I go through and make the predictions in the USAU contest with Ultiworld, I’m going to try and write about the reasoning for each pick.

Photo by Christina Schmidt of

Comments Policy: At Skyd, we value all legitimate contributions to the discussion of ultimate. However, please ensure your input is respectful. Hateful, slanderous, or disrespectful comments will be deleted. For grammatical, factual, and typographic errors, instead of leaving a comment, please e-mail our editors directly at editors [at]