Thoughts on using Double Elimination for the Top X cut

[spags]

@Kingsley

I think the current tournament system is flawed, but nowhere near as bad as the old one, which led to unpleasant fake games during at least some point of the elimination rounds in literally every tournament I went to using the system.

Not to mention certain deck types, as damage dex were at a severe disadvantage in match play.

[Arkhon]

@Squash957

First off, the generalization that preference of Elimination rounds is because I’m American is a pointless abstraction and completely unneeded. I made it clear that this was MY opinion and I am not speaking for my country as a whole.

I wasn’t suggesting that you were speaking for your country, I was implicitly suggesting that your opinion is probably influenced heavily by American culture. This includes professional sports but there’s also the classic “Ameritrash vs. Eurograme” divide, which is characterised partially by a “luck vs. skill” divide. This is nothing to take offence at, it’s just an observation.

I agree that it’s quite exciting to have big occasions with high stakes; everyone loves an underdog story and rooting for the little guy. But the underdog shouldn’t be able to win over a sustained period. If you were a professional sportsperson at the top of your game, I’m sure you’d prefer to play within a structure that rewards your skill and doesn’t give the players you should be beating a better chance to topple you.

I wasn’t specifically referring to football (soccer), but the analogy applies. Winning a national league is a big deal in most of our sports (mostly because of television and sponsorship money). But even the UEFA Champions League you referred to starts with group pools (mini leagues) and then the knockouts are decided by aggregate score over two legs. Variance and home advantage are hugely important, so we do what we can to mitigate those.

@Squash957

If you think 6-8 Swiss rounds are any less susceptible to variance than Swiss + Elimination and properly reflect a League format, I’d disagree vehemently.

Of course there’s the same level of variance in Swiss as in Elimination, but the point is that in a purely Swiss tournament that piece of good/bad fortune is less swingy. If I’m unbeaten all day and I’m going to drop a game to the guy placed 6th, I’d prefer it was in the seventh round of Swiss and then rescue a draw when it won’t damage my position too badly, rather than it eliminating me from a tournament that I’ve been totally dominating all day.

[Alexfrog]

@dashakan

Where are you getting this idea that the corp wins all the time? It seems as if you aren’t even playing the game anymore.

Where are you getting this idea that my post was meant to be serious?
Someone claims “The game is SO balanced that EVERY match between even players ends in a split”. So I respond with equal hyberbole that the corp wins 100% of the time, thats why its a split.

[dashakan]

Because I have seen it said other places in the past that you believe that, so I had reason to believe you weren’t being hyperbolic. I have even read you stopped playing because of the alleged “corp imbalance”. But it is hearsay.

[hypomodern]

@dashakan

I also disagree in general with the notion that “good players always split”

Nothing is “always” true, but splits are the most common outcome I have observed at the top tables in the two regionals I went to and the two others I watched on the 'tubes. So there is some reason to believe that a 1-1 split is a common outcome between two players of relatively equal skill piloting decks that don’t suck in both matchups.

So then we might ask whether it is consistently corp/corp or runner/runner in the win column :).

[CrimsonWraith]

@hypomodern

So then we might ask whether it is consistently corp/corp or runner/runner in the win column .

/shamelessplug

And if everyone used BGG user tobinlopes’ Netrunner tournament worksheet, we might have some idea.

For what it’s worth, the Colorado Regional swiss rounds last month had 23 Corp splits, 22 Runner splits, and 67 sweeps. It’s the only local tournament in the past four months (We have data for 7 tournaments in-state in that timeframe) in which the Corp out-performed the Runner.

Edit - For further clarification: Two of the other six tournaments were dead-even, and the Runner advantage in the remaining four tournaments was quite slim. I would heartily agree that the game is fairly well balanced between Corp and Runner right now, particularly at competitive levels. But that’s just my two cents. =)

[dashakan]

In one of the regionals I won, I lost only 2 corp games, in the other I won all my corp games and lost 3 runner games. I think the game is very even right now.

[SamRS]

Personally I think the system is by far the best version to date, and double elim has made for much more interesting brackets (read: not this Clockwork Gamer Store Championship: Finals, Sam vs. Jonathan - YouTube ). Preventing the odd playing to tiebreaker scenarios has done a lot of good as far as watching the game goes.

[spags]

Yeah, but, what do you know? Have you looked at the OCTGN data?

[Alexfrog]

Anyway, being realistic, this month on OCTGN its 56.5% corp, 43.5% runner. (Note: More imbalanced in favor of corp than the game was imbalanced in favor of runner at the release of C&C, before Jackson Howard came out).

For even matches to “always split”, this would mean that whenever a player won a game there would have to be some effect that VERY STRONGLY gave a huge advantage to the other player in the next game.

The only realistic things that I can come up with would be:

1. If the loser of game 1 is given a huge advantage in game 2, such as going first in a game where going first is a big deal. (Not applicible to netrunner).
2. If the winner and loser switch sides between two gameplay modes (corp and runner), where one of the two has a large advantage.

In a balanced format, the evenly matched player who ‘luckily’ won game 1 would have a 50% chance to win game 2 and sweep, and a 50% chance to lose and split.

[Kingsley]

[quote=“Alexfrog, post:54, topic:1395”]
Anyway, being realistic, this month on OCTGN its 56.5% corp, 43.5% runner.[/quote]

As expected.

@Alexfrog

(Note: More imbalanced in favor of corp than the game was imbalanced in favor of runner at the release of C&C, before Jackson Howard came out).

Not quite. As I mentioned earlier, when C&C came out, top-level play balance was somewhere around 75-25 runner. It seems extremely unlikely that top-level play balance is anywhere near that skewed at this stage.

Standard theory holds that runners benefit more from player skill increase than corps do, so right now top-level balance is likely closer than the average balance. I think we’re probably pretty darn close to 50-50 at the highest levels of play, though obviously we would need a more detailed analysis to confirm this.

[hypomodern]

@Kingsley

Not quite. As I mentioned earlier, when C&C came out, top-level play balance was somewhere around 75-25 runner. It seems extremely unlikely that top-level play balance is anywhere near that skewed at this stage.

Right, but Alex is talking about a different pool: the general OCTGN pool. When C&C came out this pool was 55-45 runner-corp.

If top-level play is roughly balanced, we should see about 50% splits, 50% sweeps.

[Kingsley]

@hypomodern

Right, but Alex is talking about a different pool: the general OCTGN pool. When C&C came out this pool was 55-45 runner-corp.

Yes, I understand that. My point is that the general balance is not reflective of the high-level balance, and the high-level balance is more important.

[hypomodern]

@Kingsley

My point is that the general balance is not reflective of the high-level balance

I don’t think anyone disagrees with this :). It is at any rate not incompatible with Alex’s observation that in the general pool for which we have stats, imbalance has increased (45-55 => 56-44) across the Spin Cycle.

[Kingsley]

[quote=“hypomodern, post:58, topic:1395”] It is at any rate not incompatible with Alex’s observation that in the general pool for which we have stats, imbalance has increased (45-55 => 56-44) across the Spin Cycle.
[/quote]

Yes, my point is that the general pool balance is not particularly relevant to the broader conversation, which is about using double elimination for the top X cut and how often people will split vs. sweep in the final round.

[kiv]

In Boston we had around 70 players, started around 1:15pm with a scheduled 1pm start time, and then had 6 rounds of Swiss and Top 16 which was insanely long. I left around 1am after losing my last two games terribly due to fatigue and there were still people playing. No idea when the tournament actually ended.

I would have been extremely happy if Top 16 was on a second day, since I travelled down for the weekend anyway. Or if it started at 10am. As it is I’m pretty ambivalent about going through such a slogfest again.

[hypomodern]

Ah, see: if you’d said that in the first place we could all have avoided this series of minute-long comments :).

[jacek_wieszacze]

[quote=“chrism6794, post:60, topic:1395, full:true”]Or if it started at 10am.[/quote]This is the most important thing. I’m TOing a Regionals this weekend, I expect 6 rounds and top 8, I plan to start round one at 10:30 and after reading all those reports I regret a bit that I didn’t schedule this half an hour earlier (to make up a bit for this I plan to give people possibility to preorder food at registration so when they finish round 4 it is waiting for them and I can cut the extra food break to 20 minutes thanks to this.

[hasuprotoss]

A weird occurrence in the top cut is that you can play the same person twice and play the same exact matchup both times. I had the option to do that in the loser’s finals, but decided in the end to switch it up as I was more comfortable with my runner deck than my corp deck.

I imagine this type of scenario is rather rare, but the fact that it happened seems odd.

[PeekaySK]

@jacek_wieszacze

I’m TOing a Regionals this weekend, I expect 6 rounds and top 8

After attending two regionals this year, I’m positive that if this structure stays in place for next year, I’ll simply plan tournaments with a cut as two-day events. Otherwise, the games in top 8 are a lot of things, but pinnacles of competitive play they’re not. I’ve seen people die to to a Scorch they actually saw because they forgot they wanted to draw a card and took a credit instead, I’ve seen people forget to shake a Snare tag and Kati loaded to 12 credits before either player noticed it, I’ve seen the same Psychic Field ran twice in the space of 3 turns… with the amount of sheer fatigue and exhaustion, it was like watching a zombie play with a bot, most of the time.

The two-day plan is especially viable now that we have actual, proper draft product.