2019 Terrapin advanced stats

[wcheng]

Here are links to detailed stats for each Terrapin site so far. Do not share these with anyone who does not already have access to this private forum.

2/23 Discord
3/2 Auburn
3/2 Maryland
3/2 MIT
3/2 Stanford
3/9 Toronto
3/16 MSU
4/20 UCSB
4/20 WUStL

[warum]

Will the every.buzz for this tournament eventually be available?

[wcheng]

Will the every.buzz for this tournament eventually be available?

Ophir says they are not deployed yet.

[Sen. Estes Kefauver (D-TN)]

I think that quizbowl has shown repeatedly at all levels that it can't be trusted not to cheat, and I think the existence of links that tell you literally every answer to expect in the tournament is a recipe for cheating, so I don't think it's a good idea to have this thread until the set is totally clear, regardless of whether it's behind a locked forum.

[34 + P.J. Dozier]

I think that quizbowl has shown repeatedly at all levels that it can't be trusted not to cheat, and I think the existence of links that tell you literally every answer to expect in the tournament is a recipe for cheating, so I don't think it's a good idea to have this thread until the set is totally clear, regardless of whether it's behind a locked forum.

I'm pretty sure advanced stats have been used for quite a while now for many tournaments (and they have always been made available pretty immediately in the corresponding discussion forum), and there haven't been any noticeable instances of anomalous performances on tournaments that use advanced stats, as far as I know.

[Sen. Estes Kefauver (D-TN)]

The question isn't "Has this particular [very new] technology been used yet for cheating?" but rather "Have quizbowlers proven repeatedly that if they're given the chance to easily cheat, someone will eventually cheat?" NAQT famously ignored early warnings that there was a flaw in Ginseng that allowed people to access the first 40 characters of each tossup, and we all know how that ended up. Given that nothing relies on the early release of these advanced stats besides sating our curiosity, I see no downside to NOT publishing every single answerline for a tournament in a single spreadsheet before every mirror of that event has been run.

[ryanrosenberg]

The question isn't "Has this particular [very new] technology been used yet for cheating?" but rather "Have quizbowlers proven repeatedly that if they're given the chance to easily cheat, someone will eventually cheat?" NAQT famously ignored early warnings that there was a flaw in Ginseng that allowed people to access the first 40 characters of each tossup, and we all know how that ended up. Given that nothing relies on the early release of these advanced stats besides sating our curiosity, I see no downside to NOT publishing every single answerline for a tournament in a single spreadsheet before every mirror of that event has been run.

what if i told you that hosts regularly distribute files with every single answerline and clue to teams right after the tournament

[Sen. Estes Kefauver (D-TN)]

I'm pretty sure the tournaments I've been to in the last 4 years either gave out paper packets (SCT) or else "distributed" packets via them being publicly released on the archives. I had to ask around back channels for uncleared sets, which is something predicated on a higher degree of trust than if someone were posting a link with a .zip file of a set in its private discussion forum.

[CPiGuy]

When will this be updated with advanced stats from mirrors this weekend and last weekend?