Re: ACF Nationals 2019 - 4/13-14/2019 - University of Pennsylvania
Posted: Fri Feb 15, 2019 12:50 pm
ACF has decided to expand the Nationals field to 48 teams! Emails will be going out to newly qualified teams shortly.
Sponsored by the Partnership for Academic Competition Excellence (Twitter: @PACENSC)
https://hsquizbowl.org/forums/
The ACF Nationals field has been closed at 48 teams. We will not be able to expand further due to packet limitations.ryanrosenberg wrote: ↑Fri Feb 15, 2019 12:50 pm ACF has decided to expand the Nationals field to 48 teams! Emails will be going out to newly qualified teams shortly.
Teams are limited to six players on a roster without prior permission from the TD. However, if your team has a logistical situation that would necessitate more than six players on a roster, email [email protected] explaining the circumstance and ACF will consider an exemption.
For the sake of completeness: Not only did we already know that, we already invited him and were already turned down. ACF's official position now is that Gritty is canceled.AKKOLADE wrote: ↑Sat Mar 30, 2019 7:31 pm Did you know that you can invite Gritty to make a special appearance at your event?
40th St - Greek Lady (Mediterranean), Allegro's (pizza/Italian), Qdoba, U-Town (Korean), Terakawa (ramen), Pelicana (Korean chicken)Can someone from Penn please post a list of reliable lunch places that they would recommend for a speedy, close by lunch?
The e-mail has lunch from 1 to 2 PM
This is correct; lunch will be after Round 5 so as to have lunch at a more reasonable time and not to squeeze teams that have to play tiebreakers during lunch.
Can I ask why the tiebreakers policy have been changed? This policy is a departure from previous ACF Nationals, in which all teams were given the chance to move into a higher bracket. Most teams, like my own, will probably end up not playing any tiebreakers. This policy also doesn't save any packets.ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm The format for this year's Nationals has changed some from last year, so I wanted to provide a write-up of this year's format to clear up questions teams may have, and please post any outstanding questions in this thread or message me.
Prelims
Teams will be divided into 6 groups of 8, and play a limited-single-bye round robin on submitted packets, with each team playing 7 games (Rounds 1-8). Brackets will then be split 2/2/2/2 for entry into the top, second, third, and fourth playoff tiers. At the end of the round robin, only ties that concern entry into the top playoff tier will will be broken on a tiebreaker packet (Round 9); all other ties will be broken by PPG.
Playoffs
The 12 teams in each tier will be split into two playoff brackets of six. Each team will go into the same playoff bracket as the other team from their prelim bracket. Specifically, teams from prelim brackets A, D, and E will go into one bracket per tier, and teams from prelim brackets B, C, and F will go into the other. Teams will play the four teams they have not already played (Rounds 10-13), carrying over their prelim game. Playoff brackets will then split 3/3 based on playoff record. Only ties in the top playoff tier will be broken on a tiebreaker packet (Round 14); all other ties will be broken by playoff PPG.
...
Overall, most teams will play 14 games, plus any tiebreakers. The plan is to play the prelim rounds and playoff rounds all on Saturday, but the last playoff tiebreaker may be moved to Sunday morning if Saturday is running late (the current conservative estimate has Saturday ending at 10 PM for teams in tiebreakers).
Determining which tiebreakers would need to be played and running them was a major cause of delays at ACF Nationals last year. Potentially running up to 18 tiebreakers (and realistically, at least 3-5) is a huge logistical constraint for the control room and seemed likely to continue causing delays moving forward, especially with rebracketing having a separate break this year.person361 wrote: ↑Tue Apr 09, 2019 9:44 pm Can I ask why the tiebreakers policy have been changed? This policy is a departure from previous ACF Nationals, in which all teams were given the chance to move into a higher bracket. Most teams, like my own, will probably end up not playing any tiebreakers. This policy also doesn't save any packets.
This format does make preliminary seeding slightly more meaningful, but not tremendously so -- it's essentially swapping a game in crossovers for a game carried over from prelims, and both teams will have had the same prelim finish position (i.e. a 1 seed for a 1 seed). Given that Nationals' prelim seeding has historically been fairly robust, as well as the packet constraints, we believe this is the fairest workable format.person361 wrote: ↑Tue Apr 09, 2019 9:44 pm Secondly, if I understand correctly, the reason behind teams from the same prelim bracket moving into the same playoff bracket is because not enough packets were written. But, in addition to teams playing 1 fewer game, this format seems like it would exacerbate seeding issues from the prelims, which seemed to be a contentious issue at ICT this year.
You also need a tiebreaker for entry into finals, so there would have to be three tiebreakers, and thus no emergency packet.person361 wrote: ↑Tue Apr 09, 2019 9:44 pm If teams from the same bracket are split up, there will be 16 regular rounds (8+5+3), plus 2 tiebreakers, plus 2 finals, for a total of 20 packets needed, the exact number available. The problem with this is that one of the tiebreaker rounds will have to be played on a team-submission packet. A possible solution is that if a team that wrote the tiebreaker packet is in a playoff tiebreak, then that tie alone can be broken by PPG, or alternatively the two tiebreaker rounds can be switched in the hope that the same team won't also be in the superplayoff tiebreak.
Am I reading this right that ties that could impact UG and D2 titles will not be played off unless they involve top bracket placement?ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm The format for this year's Nationals has changed some from last year, so I wanted to provide a write-up of this year's format to clear up questions teams may have, and please post any outstanding questions in this thread or message me.
Prelims
Teams will be divided into 6 groups of 8, and play a limited-single-bye round robin on submitted packets, with each team playing 7 games (Rounds 1-8). Brackets will then be split 2/2/2/2 for entry into the top, second, third, and fourth playoff tiers. At the end of the round robin, only ties that concern entry into the top playoff tier will will be broken on a tiebreaker packet (Round 9); all other ties will be broken by PPG.
Playoffs
The 12 teams in each tier will be split into two playoff brackets of six. Each team will go into the same playoff bracket as the other team from their prelim bracket. Specifically, teams from prelim brackets A, D, and E will go into one bracket per tier, and teams from prelim brackets B, C, and F will go into the other. Teams will play the four teams they have not already played (Rounds 10-13), carrying over their prelim game. Playoff brackets will then split 3/3 based on playoff record. Only ties in the top playoff tier will be broken on a tiebreaker packet (Round 14); all other ties will be broken by playoff PPG.
This is a good point, my apologies for missing it previously. All ties that impact title contention (overall, UG, or D2) will be played off.Illinois Admin wrote: ↑Tue Apr 09, 2019 10:26 pmAm I reading this right that ties that could impact UG and D2 titles will not be played off unless they involve top bracket placement?ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm The format for this year's Nationals has changed some from last year, so I wanted to provide a write-up of this year's format to clear up questions teams may have, and please post any outstanding questions in this thread or message me.
Prelims
Teams will be divided into 6 groups of 8, and play a limited-single-bye round robin on submitted packets, with each team playing 7 games (Rounds 1-8). Brackets will then be split 2/2/2/2 for entry into the top, second, third, and fourth playoff tiers. At the end of the round robin, only ties that concern entry into the top playoff tier will will be broken on a tiebreaker packet (Round 9); all other ties will be broken by PPG.
Playoffs
The 12 teams in each tier will be split into two playoff brackets of six. Each team will go into the same playoff bracket as the other team from their prelim bracket. Specifically, teams from prelim brackets A, D, and E will go into one bracket per tier, and teams from prelim brackets B, C, and F will go into the other. Teams will play the four teams they have not already played (Rounds 10-13), carrying over their prelim game. Playoff brackets will then split 3/3 based on playoff record. Only ties in the top playoff tier will be broken on a tiebreaker packet (Round 14); all other ties will be broken by playoff PPG.
This is a terrible decision. Why is the top bracket the only bracket worth playing a tiebreaker for? The difference between playoff brackets is significant not just for final placement (which I suppose you could argue matters less for weaker teams)— it also determines the quality of opponents that a team will be playing for the rest of the tournament. It is vital that rebracketization is as fair as possible so that the majority of teams can play their most suitable opponents, but here ACF is saying that only very good teams can have that right.ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm At the end of the round robin, only ties that concern entry into the top playoff tier will will be broken on a tiebreaker packet (Round 9); all other ties will be broken by PPG.
These points are far overstated; ACF certainly does not hate teams that aren't expected to content for top-bracket entry.justinfrench1728 wrote: ↑Tue Apr 09, 2019 11:21 pmThis is a terrible decision. Why is the top bracket the only bracket worth playing a tiebreaker for? The difference between playoff brackets is significant not just for final placement (which I suppose you could argue matters less for weaker teams)— it also determines the quality of opponents that a team will be playing for the rest of the tournament. It is vital that rebracketization is as fair as possible so that the majority of teams can play their most suitable opponents, but here ACF is saying that only very good teams can have that right.
Cutting vital games for teams solely because that team isn't going to the top bracket is not an acceptable way to make a tournament run faster.
Having played PACE NSC 2018, I did not notice any delays anywhere significant enough where I would say "let's cancel some games to move this along."
If PPG is so robust, then surely it should be acceptable for the top bracket as well. Clearly, ACF is giving top bracket teams priority. In many cases, this is perfectly reasonable— the difference between first and second place, or even between seventh and eighth place, is far more significant than the difference between 26th and 34th. However, placement is not the driving factor in the importance of these tiebreaker games: as mentioned above, the tiebreaker games greatly determine the quality of opponent that each team will play for the majority of the tournament! Unlike placement, this is equally important across all skill levels.
If the only concern is placing a team in the bracket that most matches its skill, PPG over seven prelim games is almost certainly going to be more accurate than the result of a single half-packet game.If PPG is so robust, then surely it should be acceptable for the top bracket as well. Clearly, ACF is giving top bracket teams priority. In many cases, this is perfectly reasonable— the difference between first and second place, or even between seventh and eighth place, is far more significant than the difference between 26th and 34th. However, placement is not the driving factor in the importance of these tiebreaker games: as mentioned above, the tiebreaker games greatly determine the quality of opponent that each team will play for the majority of the tournament! Unlike placement, this is equally important across all skill levels.
justinfrench1728 wrote: ↑Wed Apr 10, 2019 4:18 pmHaving played PACE NSC 2018, I did not notice any delays anywhere significant enough where I would say "let's cancel some games to move this along."
ACF Nationals is a hard tournament with long questions; of course there will be delays! If the delays are really that untenable, then ACF should get to the root of the problem by writing questions that are perhaps slightly shorter or slightly easier.
If PPG is so robust, then surely it should be acceptable for the top bracket as well. Clearly, ACF is giving top bracket teams priority. In many cases, this is perfectly reasonable— the difference between first and second place, or even between seventh and eighth place, is far more significant than the difference between 26th and 34th. However, placement is not the driving factor in the importance of these tiebreaker games: as mentioned above, the tiebreaker games greatly determine the quality of opponent that each team will play for the majority of the tournament! Unlike placement, this is equally important across all skill levels.
Since adjusting question length is obviously not feasible for this year, I will suggest that ACF select which tiebreakers will be played off based the margin of their PPGs rather than their expected final placement.
Goofy Evanescence Vine wrote: ↑Wed Apr 10, 2019 4:38 pmIf the only concern is placing a team in the bracket that most matches its skill, PPG over seven prelim games is almost certainly going to be more accurate than the result of a single half-packet game.If PPG is so robust, then surely it should be acceptable for the top bracket as well. Clearly, ACF is giving top bracket teams priority. In many cases, this is perfectly reasonable— the difference between first and second place, or even between seventh and eighth place, is far more significant than the difference between 26th and 34th. However, placement is not the driving factor in the importance of these tiebreaker games: as mentioned above, the tiebreaker games greatly determine the quality of opponent that each team will play for the majority of the tournament! Unlike placement, this is equally important across all skill levels.
I'm a bit confused here -- this is a new policy this year implemented in response to significant delays from last year. Additionally, all teams are guaranteed the same number of games (14), and all teams will play the same number of games unless they are involved in tiebreakers or finals.Periplus of the Erythraean Sea wrote: ↑Wed Apr 10, 2019 5:08 pm I think it's a bit strange that in a field where the middle bracket teams are stronger than ever, we're still using a schedules that privilege the best teams in the game. I think it was a really good thing that lower bracket teams got an equal number of games as upper bracket teams as last year, and I'd like to see this replicated next year, and it's disappointing that this year's logistics team is snubbing lower-bracket teams in more ways than just this one. I would similarly also suggest that ACF Nationals be stringently length-capped in the future, except for perhaps the finals packets.
Just to clarify, does "title contention" include the top 3 or just the top 1?ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm At the end of the round robin, only ties that concern title contention (overall, UG, or D2) will will be broken on a tiebreaker packet (Round 9); all other ties will be broken by PPG.
Just the top 1.person361 wrote: ↑Wed Apr 10, 2019 6:07 pmJust to clarify, does "title contention" include the top 3 or just the top 1?ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm At the end of the round robin, only ties that concern title contention (overall, UG, or D2) will will be broken on a tiebreaker packet (Round 9); all other ties will be broken by PPG.
As far as I remembered, the NSC stats program allows for near-instantaneous determination of the tiebreaker scenario after the final scoresheet goes in since you didn’t have enter add individuals scoring.Cody wrote: ↑Wed Apr 10, 2019 10:07 amThese points are far overstated; ACF certainly does not hate teams that aren't expected to content for top-bracket entry.justinfrench1728 wrote: ↑Tue Apr 09, 2019 11:21 pmThis is a terrible decision. Why is the top bracket the only bracket worth playing a tiebreaker for? The difference between playoff brackets is significant not just for final placement (which I suppose you could argue matters less for weaker teams)— it also determines the quality of opponents that a team will be playing for the rest of the tournament. It is vital that rebracketization is as fair as possible so that the majority of teams can play their most suitable opponents, but here ACF is saying that only very good teams can have that right.
Cutting vital games for teams solely because that team isn't going to the top bracket is not an acceptable way to make a tournament run faster.
Packet tiebreakers for top-bracket entry only is a defensible position balanced against the delays all-bracket entry tiebreakers engender and the general robustness of PPG as a tiebreaker. The problem with all-bracket entry tiebreakers is that they do significantly delay the tournament because there are far more tiebreaker situations to determine and read -- look no further back than PACE NSC 2018. These delays impact the quality of play and the quality of the tournament for all teams in all brackets. I get it, it sucks not to be able to break a tie at the buzzer, but that isn't even guaranteed to be the fairest result (half packets, anyone?). This decision seems like a necessary and good compromise to me, given the length of Saturdays on ACF Nationals past.
There were teams that finished playing tiebreakers something like 5 minutes before the lunch break was supposed to end at 2018 NSC. (This is a good place to mention that this is probably PACE's highest priority improvement for this year and some fixes are being implemented.) I'll concede that I should've used a better example where delays did affect all teams (since some brackets broke cleanly), but every ACF Nationals in the last 5 years has been afflicted by this problem to a lesser or greater extent, and it means that rounds extend well farther into the night than they should.Benin Rebirth Party wrote: ↑Thu Apr 11, 2019 2:05 amAs far as I remembered, the NSC stats program allows for near-instantaneous determination of the tiebreaker scenario after the final scoresheet goes in since you didn’t have enter add individuals scoring.
It seems especially concerning since ACF is removing games from last years schedule to save time, which I get, lunch at 3pm is not great. Losing three rounds seems like there would be ample time to spend figuring out tiebreaker scenarios, which would certainly take less than even a half round of quizbowl.
Could you explain what you mean here? (bolding mine.) I have read Ryan's post very carefully and do not see any difference in the schedules for top-bracket teams and non-top-bracket teams aside from the choice of tiebreaker. (Nor would one expect there to be, as 48 teams wouldn't require a schedule with asymmetry at any stage.)Periplus of the Erythraean Sea wrote: ↑Wed Apr 10, 2019 5:08 pmI think it's a bit strange that in a field where the middle bracket teams are stronger than ever, we're still using a schedules that privilege the best teams in the game. I think it was a really good thing that lower bracket teams got an equal number of games as upper bracket teams as last year, and I'd like to see this replicated next year, and it's disappointing that this year's logistics team is snubbing lower-bracket teams in more ways than just this one. I would similarly also suggest that ACF Nationals be stringently length-capped in the future, except for perhaps the finals packets.
There are superplayoffs per my understanding, they just occur for every bracket like at ICT.Periplus of the Erythraean Sea wrote: ↑Thu Apr 11, 2019 9:21 am I misread the schedule. At some point I heard something about super playoffs as well and this appears to be untrue.
I'll go out on a limb and say the word "superplayoffs" has exhausted its utility in quiz bowl, since originally it meant "a final, after-playoffs stage for the best teams" and now it just means "the thing that comes after playoffs". When writing format descriptions (e.g. for NSC) in recent years, I have almost always used the words "placement" to refer to the third stage of a three-stage format. For that matter, quiz bowl uses the word "playoffs" in a manner almost completely unlike other competitive activities.CPiGuy wrote: ↑Thu Apr 11, 2019 9:25 amThere are superplayoffs per my understanding, they just occur for every bracket like at ICT.Periplus of the Erythraean Sea wrote: ↑Thu Apr 11, 2019 9:21 am I misread the schedule. At some point I heard something about super playoffs as well and this appears to be untrue.
If the move to Sunday does occur, would it be possible to play tiebreakers for everybody? Teams could be contacted overnight about locations, and can also be asked to arrive earlier than the rest of the field.ryanrosenberg wrote: ↑Tue Apr 09, 2019 1:37 pm The plan is to play the prelim rounds and playoff rounds all on Saturday, but the last playoff tiebreaker may be moved to Sunday morning if Saturday is running late (the current conservative estimate has Saturday ending at 10 PM for teams in tiebreakers).
This has been fixed.Benin Rebirth Party wrote: ↑Sat Apr 13, 2019 12:21 am On this schedule we have two byes and play both Chicago and WUSTL in the same round.