Here are the correct seedings, along with an explanation of allowable deviations from those seedings. Mentally, I start off with last year’s results and adjust based on this year. However, I really think you’d get about the same result if you started with this year’s and adjusted based on last year. There is a remarkable consistency from year to year.
First, I put the teams into groups. Deviations are allowable within a group but not between groups.
Furious
Sockeye
DoG
Ring, Bravo, Subzero, Revolver
Condors, Rhino, Chain
Vicious
Metal, BAT
Machine, Truck Stop
Monster
Sockeye really should be first, which I found to be a big surprise to me since I weight a championship last year so highly, but they beat Furious 4-1 this year, three of those wins coming in August/September. But Furious won the Region, so there. And the two teams met in the finals of the two best tournaments and finished 1-2 last year.
At first I was going to group DoG with the teams below before putting us 3rd, but the body of work coupled with a semis appearance last year is too much. The lowlight of the season (DoG's only non-victory, in fact) was ECC, but Condors and Rhino finished just one win above DoG, and the two had easier schedules since they each got to face all of the East Coast times while DoG had to play all the West Coast teams (and in fact DoG had about the same RRI at the tournament).
[But a quick note about RRI or any other true-strength predictor. For purposes of seeding, it overweights close losses.]
Before looking at the data, I was going to put Bravo and Revolver 4/5, but neither has a strong body of work this year. Bravo's win at Colorado Cup (early August, at home) is outweighed by dismal performances at ECC and Labor Day. Revolver has done OK but not worthy of top 4, other than being NW #3 (only one tournament final, which they lost). Ring has a high RRI, but part of that is from winning in Texas in early May, hardly relevant. They did about the same as DoG at ECC, but lost twice at Chesapeake in their only other tournament, and they finished 13th at Nats last year. Subzero lost by 1 to Furious in the quarters last year, made semis at Labor Day, and got knocked out by DoG in finals of Tuneup and semis of Boston Invite. They trailed at Colorado Cup, but the body of work makes them the surprise #4 seed.
Rhino is like Revolver, but finished 4th at Regionals. Condors have a decent RRI, but finished in the middle at all of their competitive tournaments, and lost big to Bravo. Chain finished low at Nats last year, made finals of Chesapeake but lost three times, played .500 at Labor Day.
Vicious won Chesapeake in their only tournament, finished 15th at Nats last year. Lost big in the finals to Chain, but beat DW handily twice, and DW made quarters last year and probably would have earned about a 12 seed had they made it this year.
Metal and BAT finished 11/12 last year, both finished 2nd in their Region. Metal seems to have done a little better this year.
Truck Stop could conceivably be included with those guys, and maybe they should. Machine is C#3.
Monster has lost to Oaks, Illinois X, Old and In The Way, and Haymaker this year.
So, the seedings:
Furious
Sockeye
DoG
Subzero
Bravo (won CC and made quarters)
Revolver
Ring
Condors (get the nod based on last year)
Rhino
Chain
Vicious
Metal
BAT
Truck Stop
Machine
Monster
Of course, all of this may be wrong.
Subscribe to:
Post Comments (Atom)
14 comments:
Giving yourself Revolver as the second seed in your pool, Vicious third, and Truckstop fourth. Clever indeed.
Revolver is no pushover in any way.
Truckstop will be.
Jury is out on Vicious.
-Interestingly, one could last yr.'s Nationals final standings to get seedings thru ~10, at least, very similar to yours (assuming you sub in this yr.'s analogs):
Furious, Sockeye, DoG, Revolver, Bravo, SubZ, Ring, Chain, Condors --- then you have to figure where to put Rhino (RRI would put them at 8, btw. Ring and Chain)
-in general, from results and first-hand viewing, the teams from 4-10 and maybe 11 are extremely close, as big a grouping of similar good but not great teams as I can recall -- should make for a very interesting 1st 2 days
-Bravo had a good showing at the CC, but would you have thought your 5 seed would have gone 3-9 at consecutive tourneys, w/ 2 of those 3 wins being vs Metal and PBR?
I've got Chain and Rhino switched in my seedings. I've also switched Machine and Truck Stop. DoG has played both Machine and Truck Stop, so maybe you know better than I do. Chain beat Rhino at Labor Day.
I am completely honest when I say that I did not look at the pools that would be created until after I was done. The only exception may have been to slot Revolver in at 6 so that all the NW teams wouldn't be in the same two pools.
Vicious at #3 in our pool wouldn't necessarily be good. The word is that they aren't especially deep, and 1 plays 3 first round. And we don't have a lot of info on how good they are this year. With only two tournaments (well, three if you count Sectionals), you have to place a lot of weight on last year's #15 finish and a 17-9 (oslt) loss to Chain at Regionals. But what if their Chesapeake performance is their true level (per RRI, they'd be only 20 points behind Furious)?
I'm not too interested in the results of one game out of 80 (unless it was an elimination game at Nationals) in determining which of two teams to place higher. Chain did place a little bit higher at LD, but lost to Vicious at Chesapeake and played even with Doublewide this year, while Rhino had decent to good finishes at various WC tournaments.
I had a tough time with "last year's analogs" this year. Jam finished 3rd/4th, but Revolver doesn't deserve the same credit as JL would have. There was no NW #4, so how much of a dropoff can we assume? And in the Mid-Atlantic, Ring finished well done but a MA team made quarters and another finished above them.
I place a lot of weight on winning a tournament, which is why Bravo ended up 5 in my system (other main factors: consistent quarterfinalist, beat Condors handily at Regionals). If you look at the results of inter-region tournaments, of which I count 8 (Texas Shootout, Solstice, BosInv, Colorado, ECC, Chesapeake, LD, Tuneup), you have Sockeye with 2 wins, DoG with 2, Ring with 1 (but the earliest of all), Bravo with 1, Vicious with 1, and Justice League with 1.
I'm surprised no one has taken issue with SZ at #4.
And objectively, I'd have to say "3-11" instead of "4-11". While I think that it's clear from looking at all the data that DoG has to be seeded 3rd (although I'm not going to want to fight you and I won't necessarily think you're an idiot if you put us lower), I think the gap from 3 to 4 (or 3 to 7, for that matter) is less than the gap from 2 to 3. If I were _just_ a stat geek, I would say that the first three days of Nationals are just to determine the strength wildcards. But I'm a man desperately trying to play just one more game that matters before I retire, so bring it on.
I'll take issue with Subzero at #4. Bravo has beaten them twice this season handily and finished higher than them last year - it would be silly to seed Bravo one spot below SZ - but since that one spot is 4/5 it makes little difference.
I think when it comes down to it the 4-10 seeds are a mess. Also, the format of nationals will allow teams to perform to their level - so seeding is not that big a deal.
Yes, Bravo was undewhelming at Labor Day and ECC.
Kyle,
Yeah, but that last sentence is so important, and it's not just one tournament, but two. And while you beat them twice, one of your wins over SZ came at Labor Day, where it was your only win, while they made the semis in a tournament with 8 Nationals qualifiers (plus JL), albeit in a three-way-tie. So I'd call your head-to-head a wash, despite your 2-0 record against them, just like I'd say your head-to-head against Rhino is close to a wash, despite their 2-0 advantage (both defeats handily also) (they finished higher than you in 2 of 3 tournaments, but you won CC). With so many teams in the same neighborhood, it's very likely that you can do a "head-to-head" between two adjacent teams and find a seeming inconsistency.
And regarding "finished higher last year", the UPA can decree that a 5 v 6 consolation game is meaningful, but I can include other information from that day of play. Bravo lost in the quarters by 5 to the weakest semifinalist, while SZ took the champs to double game point in the quarters. I'm considering it to be equal, or close enough that it would be about the 8th tiebreaker. I'm not sure what other people do, but I make almost no distinction between 5 and 8 or between 9 and 12, but a lot between 4 and 5 or between 8 and 9.
Seedings look good to me although I don't see the distinctions between 4-10. I do find it interesting that the only teams to beat Furious or Sockeye (Rhino and Condors) are grouped in your 8-10 group. That alone would have have me bump them up.
Why does everyone focus on seedings and not final placement?
"But I'm a man desperately trying to play just one more game that matters before I retire, so bring it on."
So DoG is shooting for Pre-quarters this year?
Jim,
Bravo also beat SubZero in the power pools last year 15-9 (I think).
I tend to like teams that are placed right next to each other to be seeded according to their current year head to head. This is why I would seed Ring ahead of Bravo if they were going to be seeded next to each other.
Honestly, no team really "deserves" the 4th seed based on their season record. Which I would see as a vote for Bravo based on last year.
-Kyle
Everyone focuses on seedings because we haven't played yet. Afterwards, you play "what if" only amongst yourselves, until it's time for seedings again.
It _is_ interesting that only Rhino and Condors have beaten Sockeye and Furious yet are 8-10. Rhino with a 3rd place at Regionals would have taken Revolver's place and maybe even gone higher, but they're 4th, or even if JL had finished 3rd, maybe Rhino would be bumped up a little, but Revolver's overall play is worthy of something like a 6-7 and Rhino is lower. Condors are bumped down from going 0-3 on Day 1 at Nats last year (although it didn't stop Ring) and from losing to Bravo at Regionals by 6.
It is pretty close. It's only 66 RRI points from the highest of 3-10 (Ring) to the lowest (Condors) (predicted score 15-12.7), about the same as the difference from Sockeye to Furious (64) and a little more than from Furious to Ring (43). But factoring in last year and discounting pool play games a bit (many of the "upsets" that have happened this year have been in pool play games that didn't determine who made elimination rounds) leads to a little more spread.
Kyle,
Ok, I missed that, that counts for something, so I'm willing to say Bravo did better at Nats than SZ.
What I don't like about what you suggest is two things:
1. More and more, many of these head-to-head matchups are occurring at power tournaments where almost every game is against a Nats-level foe, so I place more importance on how a team did at the tournament than how any particular game went (hence you guys outperformed Rhino at CC despite losing to them).
2. It's not enough to be adjacent, I think you have to be tied. This also doesn't work in the case of intransitivities.
I started this blog entry intending to quantify the "algorithm" that I use, in response to something Gambler wrote on rsd, but it quickly got too messy. There are many little factors that generally don't matter but can.
Post a Comment