How Should A/B/etc. Teams Be Divided?

[cvdwightw]

This is something that was kind of bugging me when we were deciding UCI team assignments, and I think it's relevant to where some of the discussion of EFT/novice tournament teams has been/may be going.

If you're like any normal club, you have a wide variety of talent levels in the players you're taking to tournaments. Some clubs may be blessed with the opportunity to take multiple teams to tournaments. My question is, how should those teams be divided, especially for tournaments with lots of novices? I see three main schools of thought:

1. Best four players on team A, Next four on team B, etc. This is obviously a good idea for national tournaments if you think you have a shot at winning or things that qualify for national tournaments if you don't think you can split your best players and still qualify. I don't know how well it works for other tournaments, though, especially when you're trying to coax a halfway decent packet out of the novices on team C. The other issue here is that there will be a whole lot more blowouts as A teams run up the score on novices.

2. Place novices with good experienced players and the players in the middle form their own teams. This was tried by many teams for EFT West Coast. This meant that many novices had lower individual scores than they otherwise probably would have, but the games were closer than they could have been otherwise. It prevents a lot of blowouts but may make new players feel like they're not contributing.

3. Select teams more-or-less randomly. There's nothing inherently good or bad about this strategy, but it doesn't seem like many teams are using it.

Anyway, I'm interested to see what people think about this, especially since we've established more-or-less that more experienced players should be allowed to play "novice" tournaments if they're eligible. Essentially, if you've got those players, a few decent returners, and some freshmen of varying talent, how would you split them up for maximum happiness?

[Skepticism and Animal Feed]

2. Place novices with good experienced players and the players in the middle form their own teams. This was tried by many teams for EFT West Coast. This meant that many novices had lower individual scores than they otherwise probably would have, but the games were closer than they could have been otherwise. It prevents a lot of blowouts but may make new players feel like they're not contributing.

In theory, I like this one. Putting 3 freshman with a sophomore or junior, so they are not too overmatched, but at the same time can get reassuring comments like "don't worry, that will never come up again" or "these are all really easy to learn" on the spot if some scary question comes up.

In practice, it is unworkable because putting that one slightly older player on there will, for ACF Fall, trigger a full packet requirement, and that one guy isn't going to be able to write a full packet.

[No Rules Westbrook]

This is a good question, although it's somewhat theoretical in nature, because not many programs have a whole roster of people of widely differing abilities and are putting multiple teams out there. That said, there are a few occasions where this becomes an issue.

I think the only thing that should generally be avoided is sticking inexperienced players (ones who aren't known to be quite resistant or determined) on a team with people who are significantly better and more experienced than them - especially generalists, since they will just munch up tossups and make those players pretty hesitant to buzz. It's optimal for players to feel like they have to buzz if they reasonably think they know the answer, not like they're better off not buzzing since player X will get it eventually anyway and "save" them...same goes with bonuses, it's ideal for players to be encouraged to actively think through stuff and not just defer to whatever player X says.

[Mr. Kwalter]

I agree with Ryan. When newer players are placed on teams on their own, they feel a much stronger need to buzz. They become responsible for the team's success, and that small bit of desperation can be key. I have no problem with concentrating good players on the A team and putting the novices together to fend for themselves.

[Jeremy Gibbs Sampling]

I'm a big personal proponent of (1). I don't think putting one strong veteran on every side of novices is going to be inspiring enough to them to keep many of them around. If you have a lot of talent left over after assembling your A team, I guess you could try putting the rest of your veterans among your novice players.

The exception is your own tournaments; I don't believe in assembling your house teams for the express purpose of winning your tournament. It's a little different when you're hosting ACF Fall or the like, but for homegrown tourneys, your experienced people should be staffing and everyone else should be distributed pretty randomly.

I'm a big proponent of taking your novice teams to things like EFT, NAQT (NAQT bothers me less than it does many of you), and so forth, where their writing questions isn't an issue. They can definitely be eased into the writing side of things if you've got enough veterans that this is an issue.

If you're considering taking whole teams of novices to ACF Nationals or even ACF Regionals, well, you've got some heck of a freshman class.

Sadly, your team is gonna lose the interest of a lot of people over the course of a year. Trying to make permanent fixtures out of all the people who put their name down at the activities fair and thought it would be fun to go to a tournament is a futility drill.

[vandyhawk]

I've done all of these options at one point or another. Up until a couple years ago, we pretty much always made the strongest A team possible, then made a B team with whoever was left. While trying not to be too much of an obv brag post, though, two years ago I think we only lost one match all year at a non-nationals event, so last year we did option 2 at EFT, and at COTKU, we did option...4, I guess, where we made two teams as equal as possible, since only one of the eight people was a freshman. I guess for some teams, my newly coined option 4 might be pretty equivalent to option 2, although for us, having Paul and I on the same team still made it the clear 'A' team. I think the most fun was had by all with the two equal teams, since it made matches closer and more exciting with more contributions from everyone, and we were still able to finish 1st and 3rd. At our rendition of EFT, though, we had enough freshman to let them all play together on the 'B' team, while letting the returning players who weren't needed to staff play on the 'A' team. At a novice-type tournament where you have enough novices, I think it seems like a good idea to have them all play together, but on the other hand, not if it means that you'll also be fielding a juggernaut of experienced people that is crushing people by 500 points. It really depends on the people too - some would rather put up a whopping 5 ppg but be on an undefeated team, while others might like to score 30 ppg but go like 3-8. In short, my answer is: it depends.

[theMoMA]

I'm in concordance with letting freshmen fend for themselves. At EFT last weekend, a freshman named Chris got the unfortunate job of playing as our fourth player. He got a single question all day and told me he was afraid to buzz or contribute to bonuses.

For the most part, players have to learn how to play on their own. While it might be fun to ride coattails and get books or trophies or something, a player's first couple years are much more about becoming comfortable answering questions by hearing them to the end (if necessary) against appropriate competition. It doesn't do a novice much good to hear me buzzing on Loschmidt's paradox or Gautam on STACEE or Rob on Pirithous. They need to learn that the entropy of the universe is increasing is the 2nd law of thermo, that the blue glow of a nuclear reactor is Cerenkov radiation, and that strangling the Nemean lion was one of Heracles's labors.

I'm really glad that Chris stuck with it and played with us, but I can't help feeling he would have been better served for his own development playing on a different team.

[vcuEvan]

3. Select teams more-or-less randomly. There's nothing inherently good or bad about this strategy, but it doesn't seem like many teams are using it.

There are plenty of bad things about this strategy.

[evilmonkey]

Notre Dame brought 8 - 4 frosh and 4 non-frosh, and we split along those lines (and in the prelims, the freshman won two more games than the others).

[cvdwightw]

3. Select teams more-or-less randomly. There's nothing inherently good or bad about this strategy, but it doesn't seem like many teams are using it.

There are plenty of bad things about this strategy.

Other than that it isn't really a "strategy", I'm not sure what bad things you're referring to.

For full disclosure, UCI's recruiting has so far brought a total of zero interested freshmen. We let Jason and Marcus be "team captains" and pick from the remaining players. I guess that is more like Matt's option 4, where even when you try to make the teams equal one team ends up doing significantly better.

[vcuEvan]

I think no matter what your goal randomly assigning teams is a bad idea. If you want to give new players a chance to answer questions themselves put them on their own team. If you want to have the best chance at winning assemble your best team. If you want to have as many good teams as possible split up your talent evenly.

I don't see the need for randomness.

[Captain Sinico]

I think the optimal team allocation strategy depends on the what the other teams are. In my experience, novices are, if anything, more likely to be discouraged by badly losing a vast majority of games than by being on a team where they aren't able to contribute much. Therefore, I recommend assembling all-novice teams that have a good shot in the field wherever possible and placing novices on teams with more experienced players otherwise.

MaS

[Quantum Mushroom Billiard Hat]

Yet another side to this discussion is that a good upperclassman would likely be much less discouraging to play with than a good freshman. Players with little experience who play with a college veteran could see "well, he's good, but perhaps I can be that good when I'm a junior/senior." When playing with an especially good freshman, that outlook could change to "I'll never catch up with him," which would not be very encouraging.

[No Rules Westbrook]

Not that it's relevant to this discussion, but let's be very clear that in college qb anyone can "catch up" to almost anyone else (save a few very good experienced players). I can understand why a new player might think that they couldn't, but it should be hammered home from the start - that not only can they catch up, it's "criminally easy" (to use Weiner's phrase) to catch up.

[grapesmoker]

I would suggest having less-experienced players play together so that the novices feel like they're contributing rather than just riding a dinosaur's coattails. It's been my experience that this is a somewhat better arrangement.

[rylltraka]

Our team policy has always been to organize teams along option A except in national qualifiers, if only because it gives everyone an opportunity to win some games, which I think is psychologically more uplifting than having a decent PPG and repeatedly being blown out. At the last tournament we attended, both our teams finished with an equal record, which is how I would prefer it to be.

That said, our team is short on dinosaurs, and even our best players are not necessarily dominant, so there is open space for a new player to pick up some slack.

(On a side note, at the tournament (TWAIN), I won the neg prize despite being about 5th-6th on the neg list. Not upset, just curious)

[cvdwightw]

To be fair, you did win the "playoffs" neg prize (they entered my stats wrong for the last game so I had only 5 negs in those three rounds to your apparent 7). My guess is UCLA went with whatever stats they had found at that point. I don't know, UCLA has a short and non-illustrious tradition of screwing up the neg prize.