Here is where you are wrong. A (winning) player that drops from an event (after having played rounds) has two effects: (1) That player will not be playing against other (winning) players in later rounds, making those rounds easier for players with similar winning records. (2) it can screw up the assignment of "bye" rounds, because with one less player someone is getting the bye, and therefore not getting the chance to score any points.

I understand that you are viscerally opposed to my suggestion, but I don't think you've thought it through using the lens of a large event with 16, 32, or 64 players. I grok the reactions if you are mostly familiar with 3-round events with a small number of players, but for larger events there are essentially two scenarios we should be able to agree don't matter:

(1) Players with nothing but wins <- Should such a player drop, its not much difference as everybody moves up. Sure there can be some of the past opponents of such players which now *may* have both a tied W/L record and higher points than some folks before the drop, but we aren't trying to motivate a game where people go N-and-0 by winning N rolloffs? (note they still have more rounds to try to win with this strategy, should they wish)

(2) Players with nothing but losses <- They are already on the bottom, so there is no movement, and no one else's record changes.

In the case of all other players, if a player with "mostly wins" drops, then the gap between the top and the bottom widens. This essentially makes the event meaningless for those below the dropping player. One person dropping from the middle of the pack shouldn't be an incentive for players with worse records to drop as well, otherwise just make the event single-elimination.
If a player with "mostly losses" drops (see the example posted above, but pretend that Y dropped after round 3), all it does is tighten the race. I realize some folks see this as given 'losers' the chance to "jump the line", but all it did was put the spots after first-place into competition (in that example, for a player that legitimately won half his games and still managed to score huge points in his 'losses')

If the round-by-round data for a 64-player single event really is available, I'll be happy to look at it, and followup with my analysis. I only did the example above because Sassamo put forward an example.


A couple of other points: This has nothing to do with 'cheating', as the assumption is that this event rule would have been posted ahead of time.

Fair enough, as the reason I don't play in certain types of events is because I know the event rules in advance. But may I ask what it is specifically that is so upsetting, given that there will still be rounds to play?

[what follows is me trying to understand the opposing point of view]

I'm thinking that the main point of contention is that somehow a 'loser' is ending up higher in the rankings (e.g. second versus fifth place). We can explore this with Sassamo's example:

In the 16-player (4-round) example, after the drop but before the final round there were:

2 Undefeateds (3-0), and 7 (2-1, one is "X"). 6 players have a record that is (1-2, or worse)

After the final round, there is:

1 Undefeated (4-0)
4 players (3-1)
3 Players (2-2)
and the last player is either (3-1) or (2-2).

In the scenario where "X" is this awesome player who racks up a lot of points but somehow still loses almost every game, they will end up 2-2 (but with a lot of points), which would place them 5th out of 16, instead of finishing in the bottom half (1-3) with a boatload of points. Coincidentally, the 1-3 record would be the same as the "Beginner" who will get the bye (a win and no points).

If they win their last round, that player would finish 2nd. (because they somehow get almost all opponent's points in all games, whereas other players didn't come close, despite winning). As is often the case in events like this, 1st and 2nd won't have played against each other.

I'm trying to be fair to the opposing point of view, so I want to understand where the specific unfairness comes in.

Direct link to Nationals: http://win.wizkids.com/match.php?EVENT_ID=171884

Also, make sure to time how long it takes you to correct the past records. Then remember you're doing this in the privacy of your own home and not with those 60 some odd players waiting for pairings.

Also pay special attention to how it affects the top of the rankings. No one is arguing it's unfair to the lower half of the field (except philosophically by taking time and complicating things) but it is punitive to the top of the field that have done nothing wrong.

Posting this before I see any reply to my specific post. So this may be redundant.

The unfairness comes in two simple places: 1) at a big tournament where you have a "Top X" cut before single elimination and 2) strength of schedule.

If you lose in the first round, you will likely play against weaker opponents. If you beat those weaker opponents, you are getting a free win that can drastically change your standings in the overall tournament. A 2-1 record can easily be a 1-2 record if you win your first round; yet you're changing a record with a loss into an undefeated record.

Also note, if your suggestion is meant for larger tournaments, larger tournaments rarely have 1 person drop. So it's not a matter of 2 people getting a bump. It's a matter of possibly 4 or more. If two of those people played one of the players dropping, they could literally go from the bottom of the ranking to the top.

Thanks for recognizing this. This is the main motivation for my proposal, specifically in the context of large events in which most player's won't play against each other.

I'm not trying for a 'perfect' system for events of all sizes, I'm strictly thinking about the big events (2^N players, with N rounds of play, N>3).

I'm going to set aside the 'strength of schedule' concept. I recognize it, but it is too hard to quantify. For example, in Sassamo's example, how is that a player can almost sweep every other player's team yet still lose every game?

The examples you site must be at least a 4-round event, like Sassamo's example.

A 2-1 player (lost first round, sweeping through the losers bracket) is still going to be in the top half of players, changing the single loss to a win just means that they are playing against players with similar records (undefeated, or players with a loss but with similar number of points).

You are dead on here, where a blanket (losses to drops become wins) starts to get 'unfair':

Let's consider a big event (6 rounds, 64 players) with a cut. (what's the true upper limit in practice these days?)

I completely understand that more people drop than just one in big events. My experience has been that the more players drop, the less likely it is for players beneath them to qualify for the 'cut'. Essentially, if you can get enough players to drop, all of the players below them should also drop, since they are mathematically eliminated from play.

It seems horrific that a player 0-5 could shoot up to the undefeateds (joining at least two 5-0, and passing as many as ten 4-1) and would then be guaranteed to finish no worse than 8th, since usually in HeroClix the "top 8 is a thing". Mathematically a limit has to be put on the number of Losses to Drops that would become wins, or the final cut has to be tighter (top 4). Note that this would have to be 'worst case' where the players 5-0 & 4-1 & 3-2 after 5-rounds weren't benefiting from one of those drops as well.

In a 64-player cut with a top 8 cut (based on W-L), once you have 2 losses you are mathematically eliminated, so maybe instead we should just start dropping players once they have two losses? Is that a better solution than what I originally proposed? Would that be more acceptable to the crowd?

No, what would be acceptable is actually running swiss tournaments as swiss tournaments. What you're talking about there is double elimination. Which is fine, but that's a completely different kind of tournament. And no, you are not "eliminated" after 2 losses. Chances of making the top 8 aside, most big events like that pay out past the top 8. On top of that, you can't just drop players from an event. That's their choice, not yours. Until the cut, they have the right to play every single round, win or lose.

Its interesting that you imply I'm only familiar with small, 3-round events, when I specifically said earlier that the impact of your suggestion is MORE SEVERE ON BIGGER EVENTS. I've been involved in swiss tournaments with more than 600 people.

Its pretty obvious at this point that you have no interest in actually listening to anybody here. Swiss tournaments are the most common type of tournament format for a reason. Hate to break it to you, but you're not going to improve on it with your misguided ideas of "fairness".

This is a good point (about the top tier, I don't think anyone left in the event really has done anything wrong). Given the standard sort of event:

2^N players, N rounds, top 2^(N-K) make the cut

Assuming you still want as many players to stay in for all N rounds, to avoid the idea of punishment (for repeat winners that don't drop), you may have to limit the number of 'drops become wins' to K-1. (K-1 is the number of losses you would have to be mathematically eliminated).

I miswrote about 2 losses eliminating in the example of the 64, since it would be the case that 1 of the 15 players with a 4-2 record would finish in the top 8. So make that "eliminate players with 3 losses?".

Are you pairing players by W-L record or by Points accumulated? If W-L is used, then there is a point at which enough losses has a player mathematically eliminated from finishing at whatever arbitrary point.

Fairest would be a round-robin, but I don't think anyone has the stamina for that in HeroClix!

At this point, I am almost certain he's trolling us.

He's wrong. Straight up. Several of us have pointed out reasons why he is wrong. If he chooses to continue this avenue of discussion, that's his choice, but I would highly recommend that the rest of us no longer contribute. There really is nothing more left to say. Regardless of what the rest of you do, I'm out. Enjoy your thread.

Honestly, I'm not trolling, but responses like "You're wrong and stupid" or "I'm insulted by the suggestion" or "I hate cheaters" aren't really pointing out any reason(s) why I'm wrong. Wolverazio has been patient, and BertieWooster at least recognized that dropping out causes problems for events.

I suppose the problem I'm considering is one that others don't see as a problem: players giving losses early but then dropping from the event.
My original suggestion was obviously too radical, but it was motivated by a sincere intention to given players that lost early to the quitters an incentive to stay in the events.

I agree that no one left has done anything wrong. What I'm attempting to point out (and I would hope your continued "fixes" are beginning to illustrate) is that the people who dropped haven't necessarily done anything wrong either. So what you're doing (as Lantern points out) is basically saying, "I don't like Swiss tournaments."

Swiss (to my knowledge, I'm by no means an expert) is partially set up to account for these gaps without doing a full Round Robin. By introducing these fixes, you're making it less of a Swiss tournament without making it more of a Round Robin. Which is part of why people are getting frustrated. You're basically making it more complicated to do the very thing that Swiss already does.

Also note, when I reference 2-1 to 3-0, it's because I'm presuming you would get the most drops after the third round in a 4 round tournament, assuming a cut. Also note, there are always people that are 3-0 that eeked out a victory. Sassamo's example of a person almost sweeping while still losing is not necessary to unbalance your system, so don't fixate on it too much. It is very possible, in your system, to get shut out round 1, and dominated rounds 2 and 3, and then end up at the bottom of the undefeated (pushing out players that scored in every match and still went 2-1). Hell, as I pointed out, it's possible to get shut out in two rounds and do that.

What I find curious is that you've gone from "we should penalize people that drop" to "okay, we'll have to cap the wins to losses or force a drop."

If we just run Swiss, none of this comes up. People get the wins they get and go from there.

I would just like to point out, I've argued with Tidge on another matter before and this is just how he is when he strongly believes in his point. I don't believe he's trolling, I just think he's starting from a different place and not thinking of the implications.

It's honestly probably still better for some people not to contribute (because we might not be able to bridge that gap). But I don't want to throw rotten vegetables yet.

No one's called you stupid. Pretty much everyone has called you wrong, and tried to explain why. I am insulted at your insistence that its ok for a Judge to intentionally undermine the integrity of a tournament. I don't know what cheaters have to do with anything, but everyone hates them. Or if they don't, well, then they're probably the cheater in question.

As stated earlier, your example argument was a hand-picked scenario in which your idea worked without really screwing things up. That's called confirmation bias. However, in the vast majority of situations, especially for larger events, it severely messes things up. And as I pointed out, the swiss structure already accounts for people dropping by passing down the rankings. That's as fair as it can possibly be.

You keep insisting that a player dropping is somehow unfair to the other players in the tournament. It isn't. That's really all there is to it. You're trying to create a solution for a problem that doesn't exist, and your proposed solution causes far more problems than it "solves". It ain't broke, stop trying to break it.

A win is a win, and a loss is a loss. That's all that matters in a tournament. As long as everyone's playing the game within the rules, then results stand. That's fair.

But again, like I said a while ago, what you choose to do at events you're running is your business. Well, and the store owners business. Literally. I can't imagine they'll be thrilled at losing players and sales because of this. And that's the reality of what will happen. Because even if it doesn't change the outcome of the tournament or the rankings at the end, all that's going to matter to players is that a Judge was changing match results. That gives the appearance of impropriety. And that's all it takes.

I'd advise you to go to your game store and run this by some people that have experience running events for other games. See how your local MTG Judges/TO's react. Or your store owner/employee that runs events. They'll probably be able to get things across to you better in person than we've been able to here. And please, no sarcasm intended, let me know how that goes.

For the record, that wasn't my example, but another user's.

So Lex Luthor and Killer Moth are pals. Lex happens to play Killer Moth in round 1; Killer Moth wins. Lex goes on to win his next two rounds, while Killer Moth loses his next two rounds.

With the benefit of two extra rounds of knowledge, the pair now knows that they'd be in a better position if Lex had won that first round. So Killer Moth drops out to give his buddy a great shot at a top finish.

Killer Moth could have dropped out in the first round to give Lex a free win. But without knowing the results of rounds 2 and 3, why would he? Now he gets to delay that decision until he gets to see how Lex is doing in the tournament and give him a free win after he knows which of them has the better record.

Even if the proposed system has some merit (which I have yet to be convinced of), it seems to me that the potential for this sort of abuse kills it on sight.

This is one obvious example.

I'm just trying to figure out, though, who would want to have to go back through the tournament pairings after round 2 or 3 and people start dropping?

That's a lot of extra work and for what?

To tell someone that went 3-1, losing in round 4 against the players that kept on winning games against live opponents that they missed the cut because the guy that lost his first two games and managed to win their last two games against softer competition got retroactive free wins because another player or two decided to go do something else?

Someone quits? Oh well. That changes the fact that you lost to that player how?