preface: I'm somewhat new to the game and this was my first gencon. I'm only looking for information and rationale on the tiebreaker in joust.

Why is the first tiebreaker for joust head to head and not strength of schedule? I ask because there were a mess of people tied at 4-2 (20 pts) at gencon and i was surprised by who from that group made the cut. My buddy james and i were in that tied group and ended up missing the cut (I got 17th and he got 18th). He had an incredible SoS (including two or three 5-1 opponents) and I had opponents with losing records and 1 that dropped out completely, but because I randomly beat one person in the 4-2 group in the middle of the tourney and he didn't, I and 4-5 other players were seeded above him. He clearly faced better competition, but due to luck he was on the outside looking in. I thought for sure he'd be in and I'd be at the bottom of the 4-2 pack since i was under the false assumption that SoS was the first tiebreaker.

My thought after all this is: if the goal is to have the players who played the best during the day make the cut, the head to head tiebreaker might not be the most accurate with such a big field (i think there were 89 players). James faced significantly harder competition than me, had the same record, but was seeded below me. That just doesn't feel right to me.

Like I said before, my experience is minimal and I'm looking forward to hearing why I'm wrong

Also, when calculating head to head with such a large group of tied players (i think there were at least 10 of us), are only wins tracked, or does win-loss ratio factor in? For example, If player A is 1-0 against the tied group and player B is 1-1 against the tied group, would Player A be seeded higher, or would it go to the second tiebreaker (SoS)? If player C is 1-2 against the tied group and player D is 0-0, is player C seeded higher because of the one win? /brain hurt

Next time go 5-1 and it wont matter

from the official tourney rules:

First check if the players with the tie have played
against each other.

one (or at least such like me, who is not a native speaker) could read this as: if there are players in the tie-pool, that have not faced each other, this rule cannot be applied. moving on to the next tie breaker.

i also think that strength of schedule makes much more sense as the first tiebreaker.

btw: there is no next criterium after the SoS in the tourney rules. so what happens when after calculating SoS players are still tied?

We looked into this after this was an issue at KublaCon this year. This is what we could figure for the swiss tournament format.

The first tiebreaker after SoS is head to head among players who are tied. This usually doesn't work in a large tournament because chances are those players didn't play each other.

The next step is to do a weighted score for each player's tournament. Essentially the way it works is like this…you score points for each round you win, and you continue to score those points in the subsequent rounds. This gives an emphasis to players who win their earlier matches. For example…

In a four round tournament, player A wins their first match (3 points), wins their second match (3 points, plus 3 points from the previous win to gain 6 points this round), wins their third match (3 + 3 + 3 = 9 points this round), and loses their fourth (3 + 3 + 3 + 0 = 9 points this round) giving them a total weighted score of 27 points.

Player B wins the first round (3), loses the second round (3), wins the third round (6), and wins the fourth round (9), giving them a total weighted score of 21 points.

Player A gets the tiebreaker with the higher weighted score of 27 over Player B's 21 because they won their matches earlier in the tournament.

This is the system that chess tournaments using the Swiss tournament format use, and it's what i'll use going forward in any tournaments I run. Hope this helps.

Joey Kreins

This is of course the best option only because FFG hasn't provided a sanctioned rating system. If we had sanctioned ratings, you could get into tiebreakers such as true strength of schedule, but that's not an option for us.

Tourney of Stahleck said:

Otherwise, the tie-breaker would really need to specify that it only applied when exactly 2 people are tied.

The reasoning for head-to-head being the first tie-breaker is this (take it as you will):

• Player with 4-2 record and higher strength of schedule: "Two of the guys I beat went 5-1. One of my 2 losses was to the guy in first place."
• Player with 4-2 record and lower strength of schedule: "Yeah. And your other loss was to me."

So, who played the better tournament? Does the guy with the lower strength of schedule really deserve the cut less than a guy he beat?

Tourney of Stahleck said:

And since the tie-breakers are worded and applied as a "who gets preference" rather than "who is moved to the bottom," there is no "penalty" for head-to-head losses. Losing the head-to-head tie breaker does not automatically place you lower than players who did not play anyone else in the pool head-to-head. You just move on to SoS.

Tourney of Stahleck said:

In my experience, it is pretty rare that you need more than 2 tie-breakers, particularly as you add more swiss rounds. I think I've needed a third tie-break just one time in nearly 10 years of judging AGoT events. Ultimately, however, since FFG doesn't specify a third tie-break, the TO picks one.

Joey's method of "weighted score" (which is also called "progress") is certainly a good one (it effectively factors in the swiss "bracket" you are playing in). And while I agree that SoS first, head-to-head second, then something like progress 3rd is a good scheme (and probably easier to implement), until FFG changes their tournament rules, it is not the way AGoT events run by FFG will be scored. They will do head-to-head first.

I do think that having one more tiebreaker past the SoS would be good to specify for FFG, as this has come up in at least two major tournaments this year now…so either it's a major statistical anomaly, or it's a clear need. I think that with one additional level of tiebreak, you would probably cover 99.99% of tournaments. At that point, you leave it to TO's discretion.

Which was the second one? Kubla and what?

I think it is just silly not to have SoS as the first tiebreaker. I know I sound like a broken record, but for OP Magic has it down…they have 20 years of experience and know what will work best and what won't.

Also, I was very surprised at the 6 rounds. With over 64 didn't we need to go to 7? Isn't that in the rules or not (I seriously don't know)?

Overall I think things went VERY well, and I am not complaining since I got right where I wanted to be (4-2, and could go eat/drink). But it would be nice to have the best situation possible, and I give more credit to someone who played tough players all day.

rings said:

The decision for 6 rounds (of 50 minutes) came from the FFG "powers that be" (ie, not Nate and Damon) based on time constraints. Specifically, they needed the table space for the Netrunner event in the evening.

phoenixember said:

I think Ktom was TO at the 2nd "Trials of the Mad King" (Or did that call that one the Murder of Crows b/c it paired so nicely with the booksigning for the release of aFfC?), so perhaps he can flesh out (or totally gainsay) my recollection. I ended up being 9th by what I think was the fourth (maybe fifth?) tiebreaker, (And I don't even remember what that was), just behind Mike "cardinal23" Docherty.

I probably wouldn't "remember it to this day" if the remaining door prizes for the field of 32 weren't given away at random (outside the Top 8 and the "Sansa" prize). The old Madison crew really put together a nice spread of prizes to pick from (signed novels and unopened packs), and my finish got me the 29th prize. I shouldn't be complaining since I actually got something, but whining is in my nature, or at least my name. ("But I was going into Tosche Station to pick up some power converters!")


#Can'tLetItGo

Some things to note with standard head to head tie-breaking.

If 8th seed beats 9th seed, 9th seed beats 10th seed, but 10th seed beats 8th seed, 10th seed will be seeded highest as his head to head is the best. He lost to the player 1 seed above him, but beat the player 2 seeds above him. Then you would have the head to head between 8th and 9th and 8th would win that coming in as the 9th seed. This is pretty standard.

Basically you start at the top and work your way down. Who beat seed 1 that tied with seed 1? They're ranked higher than seed 1. Who beat seed 2 that tied with seed 2? They're ranked higher than seed 2. Who beat seed 3 that tied with seed 3? They're ranked higher than seed 3. Etc. Etc.

With regards to tournament size, tournaments with 65 - 128 competitors require 6 - 7 rounds to appropriately result in a single winner (mimic a single elimination bracket). This means you could easily do 6 rounds and cut to top 2 if you wanted to do so. Most swiss tournaments choose to cut to a larger number.

Also, Phoenixember's method was one of our tie-breaking conditions at the NYC regionals. I think it was the fourth. We went something like head-to-head, SoS, Opponent's *** SoS, Cumulative score by round, play-in.

ktom said:

well, to a certain extent it even does exactly that. the next sentence says:

If they have, the player that won
that match gets the tie breaker.

it says "player", singular, so there is the assumption of having only 2 players involved from the very start.

and in this case it even makes sense: if 2 players are tied, the direct match up can break the tie.

(of course the situation could arise where AFTER SoS there are still 3+ players tied, and 2 of them had played each other)

ktom said:

i would definately say YES. why? part of the nature of a game like AGOT is stone-paper-scissors match-ups. while having the SoS applied first, you are looking at the whole tournament. when looking at a single match first, stone-paper-scissors is mcuh more relevant - which is not reflecting the skill of players necessarily.

ktom said:

either having a last one (this could be alredy the 3rd) like "roll dice" or at least having a statement in the Tourney Rules like: more tie breakers are possible at the discretion of the TO, but must be clear before the tournament.

right now there is nothing, so that means, in case of a tie a decision has to be made by the TO anyway. all i want to say here is, that i like rules that are complete ;-)

ktom said:

well, it does happen not that rarely i think. maybe not always when it is a bout making the cut or not - fortunately. just a few weeks ago i was also in that situation: COC LCG Regional in munich. i was tied with another player after SoS (we didnt play preliminary). it was about 5th place, so no big problem, and we sponateously decided to play it out. BUT: it could as well have been for place 4. or if we both make the cut, who is seeded 5th and who 6th (whcih couold be very important because of the opponent waiting).

but thats not the point. the point is: it DOES happen, and i want to be prepared.

ktom said:

i really like this "progress" method as a third tie breaker.

mdc273 said:

But we're in a situation where the player are not ranked prior to the tournament. So it's not 8th seed beats 9th seed beats 10th seed beats 10th seed, it's player A beats player B beats player C beats player A. In that case, A, B and C all have an equal record (1 win, 1 loss) so head-to-head doesn't break the tie and we move on to the next tiebreaker.