Switch Theme:

Need to know how pairings are done.  [RSS] Share on facebook Share on Twitter Submit to Reddit
»
Author Message
Advert


Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you:
  • No adverts like this in the forums anymore.
  • Times and dates in your local timezone.
  • Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance.
  • Email notifications for threads you want to watch closely.
  • Being a part of the oldest wargaming community on the net.
If you are already a member then feel free to login now.




Made in us
Discriminating Deathmark Assassin




Out of my Mind

Have a 'Matched Play' Tournament coming up and I'm trying to figure out how pairings are done. The GT21 booklet failed to provide this information again, so I'm asking any TO's how they handle it.

After playing the mission, players go through the process of filling out the scorecard to determine the winner. What happens next? The Winners go into their own bracket and then... you randomly pair them off with each other like round 1? With no Global metric of true secondary objectives how do you pair off the winners? What system is used to determine a ranking among those in round 2 to pair players off? What about the final round?


Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in us
Never Forget Isstvan!






First round is usually random, although we usually try to not match up players that are on the same team or from the same group.

After that its winners vs winners and loosers vs. loosers with closest VP totals matched up.

Winner with 99 points will play winner with 100 points.

Looser with 23 pts will play looser with 25 points.

In the case of an odd number of players with the same points you will look at Strength of Schedual, which is the average points of the players they played against in previous rounds.


BCP does all this automatically if you set it up and use it.




ADDON: If your running your own tournament however, feel free to match people up however you like. Another option is to just go straight off VP (which is in order to put more experienced players vs each other, but this encourages submarining), or to go straight off W/L record and keep things random (but this will make some players feel cheated or encourage armies that are designed to deny points and score little themselves).

This message was edited 1 time. Last update was at 2021/06/25 19:23:24


JOIN MY CRUSADE and gain 4000 RT points!
http://www.eternalcrusade.com/account/sign-up/?ref_code=EC-PLCIKYCABW8PG 
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

 Eihnlazer wrote:
After that its winners vs winners and losers vs. losers with closest VP totals matched up.

This is where the confusion lies. The GT20 and GT21 pack ONLY tells us that the winner of a Mission is the player who scores the most VP's. Nothing else. Maybe I was just hoping that someone would tell me what is working without any instructions.

Events using VP's for anything beyond determining who the winner of an individual game is the biggest issue I have with Competitive Matched Play. We already learned back in 3rd ed. that using VP's to rank participants overall fails to work, even when trying to apply it to a far more successful BP Scoring system. Late 3rd ed, leading into 4th we saw a complete removal of Slaughter/Massive Victory/Minor Victory points from Competitive 40k because they didn't accurately reflect a player's skill. Going off Wins->VP's doesn't pair off the best players between rounds to find 'the best'.

The problem then comes back to the lack of a Tournament mission pack. The mission itself is a great primer for those who want to move into a competitive format. The Dual Objective nature teaches players to focus on scoring objectives, while providing a direction to expand an army in a specific direction. The lack of global secondary's that can be used as a measuring stick leaves a void in Competitive Matched Play where every player is equal with everyone with the same record. ie. All of the 3-0 players are the same regardless of VPs, so how do we determine the overall victor?

The removal of 'Sudden Death' in the mission makes sense as a primer as well, since it forces a player to focus on the common Primary Objective, and allows them to adjust their lists according to the Second Primary Objectives, even after they've tabled their opponent. In a Tournament setting, the only thing that 'Sudden Death' manages to accomplish is allowing a player behind in VP's the opportunity to get ahead in VP's after they've tabled their opponent. While this is possible, I can't see this happening in the 'bulk of the curve'. Once the player has managed to get 1 VP ahead, he's won and there is no longer a need to keep score.

 Eihnlazer wrote:
ADDON: If your running your own tournament however, feel free to match people up however you like. Another option is to just go straight off VP (which is in order to put more experienced players vs each other, but this encourages submarining), or to go straight off W/L record and keep things random (but this will make some players feel cheated or encourage armies that are designed to deny points and score little themselves).
Straight VP's simply doesn't work. If that were to happen, then we should go back to 3rd. Ed VP's where it's based purely off of what a player actually kills. The only reason to do this is to give GW a solid stream of data so they can balance the points cost of units, and not for any competitive reason.

I honestly don't understand the last sentence. IF VP's are used to rank players, then anyone playing an army designed to deny objectives penalizes both players anyways. In order for that statement to be true then two equally skilled players who have the unfortunate luck of being paired off in round 1, also screw each other over, as their score will be much closer to each other. Hardly seems fair when a skilled player scores top VP's in round 1 against an unskilled player, then never gets paired off against an equally skilled player as a result of it.

There has got to be a better system.

Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in us
Humming Great Unclean One of Nurgle






Pretty sure everyone involved would be keen to hear a better option if you have one. In the meantime it is a system which has worked quite well across a huge number of competitions.

Edit: On second reading, it seems you simply want to vent so nevermind me.

This message was edited 1 time. Last update was at 2021/06/28 22:20:23


Road to Renown! It's like classic Path to Glory, but repaired, remastered, expanded! https://www.dakkadakka.com/dakkaforum/posts/list/778170.page

I chose an avatar I feel best represents the quality of my post history.

I try to view Warhammer as more of a toolbox with examples than fully complete games. 
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

 NinthMusketeer wrote:
Pretty sure everyone involved would be keen to hear a better option if you have one. In the meantime it is a system which has worked quite well across a huge number of competitions.
It's difficult to not sound like I'm venting. I'm honestly trying to figure out how TO's are making this scoring system even remotely competitive. The BP system, while not perfect, was a much better system compared to the current one. I'm not opposed to the replacement of the BP system, but it should've been for an improvement to more accurately represent a competitive environment. The current system honestly feels like all the problems of 3rd ed scoring are brought back, and then they made it worse by removing the BP base.

There are no simply no instructions on what to do with the results of the games in an event. If VP's are used to determine rankings, then they should make that clear for all participating. It becomes more comparable to Golf, where the player with the 'Best' score is declared the winner. This isn't Golf. Your opponent is actively trying to deny you VP's. Golfers are trying to take holes, or remove their ability to PAR by getting their first.

IF there isn't a functional alternative, then we can start looking for a competitive scoring system that works. If that's not the BP system, then there has to be a better alternative. If not, then we need to accept the current system for the failure it is and go back to modifying the BP system and bring competition back into 40k.

What's has and has not been working in the various competitions is a completely different topic. If the VP system is what is used, then the system only works for those who still attend. Nothing is available for those who are waiting to attend.

This message was edited 2 times. Last update was at 2021/06/28 23:45:18


Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in gb
Longtime Dakkanaut




The "ideal" tournament is one where you play enough games to end up with a single undefeated player at the end. In that scenario the primary factor used to determine match-ups is simply wins/losses, with VP acting as a tiebreaker. Most tournaments don't have the time to run the number of rounds required to get a single undefeated player so you end up with a Swiss pairing system, with VPs as tiebreakers. It's not perfect, but it works well enough.

Alternatively you could have a Swiss-paired series of games, followed by a single-elimination round of games for the top players, but that doesn't work so well because of the time it takes to play a game of 40k. That's the biggest problem most 40k tournaments have. It takes so long to play a game of 40k you can realistically get 5-6 games in over a 2-day event. Many other wargames can do that amount in a single day.

This is a subject that has been debated a lot over the years, without too many alternatives being successfully implemented. If you have an alternative by all means try it, but the current system works well enough, even with its flaws.
   
Made in us
Humming Great Unclean One of Nurgle






I find it interesting that the inadequacies of the matching system are more than a footnote. Sure it is flawed, but such are rendered trivial by the matchup flaws created by the game itself. Between 40k's rule design and point costs the inconsistencies created by a pairing system amount to a cracked windshield while the engine is on fire.

Road to Renown! It's like classic Path to Glory, but repaired, remastered, expanded! https://www.dakkadakka.com/dakkaforum/posts/list/778170.page

I chose an avatar I feel best represents the quality of my post history.

I try to view Warhammer as more of a toolbox with examples than fully complete games. 
   
Made in us
Fixture of Dakka





There's no good answer for tiebreakers because they're arbitrary by definition. You're taking two players who haven't played one another and deciding which one is better despite having the same success on the day. There's no satisfying answer there and truthfully, I think players put too much emphasis on whether they got second or 4th rather than appreciate that a single loss is a successful performance.

Players often like VP based tiebreakers because they feel like they have control of their results. There's a decent argument that player perception might be more valuable than actual fairness in an inherently unfair system, but its got some significant flaws.

First off, one of the biggest problems is that it tends to show off player disparity in individual games more than serve as a way to compare similar players. What you often see is lower scores in games between higher ranked players. The high scoring games are usually from games where a new player got stomped, and ultimately, are more of a sign that a player had an easy road to the top.

I think the bigger issue is player controlled tiebreakers have a huge effect on game balance and will often have weird warping effects on the meta. Whole playstyles will get hated out simply because their route to victory isn't scored well when they lose. This is harmful in a way that really scares me off, even when it general seems to make the players who care about this sort of thing happy.

Strength of Schedule remains the best, bad option, IMO. It's main weaknesses are that it feels completely arbitrary to players and heavily assumes that players will do their best even after they've lost. It's just the only system where I've consistently looked at the standings and felt they were a pretty solid listing of the results of the games I judged.
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

Thanks for your Input @Lunarsol. I feel like you've shed some light on my frustrations and I'll attempt to respond.

 LunarSol wrote:
There's no good answer for tiebreakers because they're arbitrary by definition...
...Players often like VP based tiebreakers because they feel like they have control of their results.

This is HUGE part of why 'Matched Play' is the worst setup as far as scoring goes. VP's aren't being used as tie-breakers, they're straight up being used for ranking. The whole object of a 'Matched Play' Event appears to be scoring the most VPs, which would make the overall winner the player who scores X/300 (or however many rounds there are x100). Without instructions, you don't even have to win games as long as you're scoring VP's. I do admit that it's not going to be common that someone who loses any of their rounds will score the most VP's in an event, but it's possible. On other threads players have mentioned that most events appear to rank the undefeated players based on VP totals at the end of the event. Doesn't matter for the reason I listed above.

 LunarSol wrote:
One of the biggest problems is that it tends to show off player disparity in individual games more than serve as a way to compare similar players. What you often see is lower scores in games between higher ranked players. The high scoring games are usually from games where a new player got stomped, and ultimately, are more of a sign that a player had an easy road to the top.
This is exactly what we learned back in early 3rd edition. If the point of a tournament is to 'find the best player', then how does the current scoring system reflect that? For a group a players that fought to remove random elements, they seem to reward/penalize players for having favorable/challenging matchups.

 LunarSol wrote:
I think the bigger issue is player controlled tiebreakers have a huge effect on game balance and will often have weird warping effects on the meta. Whole playstyles will get hated out simply because their route to victory isn't scored well when they lose. This is harmful in a way that really scares me off, even when it general seems to make the players who care about this sort of thing happy.
I'm not sure I entirely follow you on this part. The dual-primary nature of the current Secondary objectives trick players into thinking they have control over their score which doesn't just affect the balance and warp the Meta IMO, it dictates WHAT players are supposed to bring. This pushes players more toward 'Netlisting' than away from it.To be successful, you need to find the best list that can tackle the 'Primary' mission objective. Once you've got that, then it's simply a matter of picking the optimal 'Secondary' to fit that list. Sure, you can pick some Secondary's to build a themed army but that might not always be the most successful. I still feel that the number of players who have both of these elements on the same list to be very small. On a funny note, I find it mildly humorous that 'Matched Play' players are upset by specialized lists, like Charadon Books, being legal in Tournaments will upset the balance when that's what they're pushing for. 'Matched Play' is the result of players telling everyone how armies should be represented. Specialized lists are doing the same thing they did to general 40k players by telling them how they want their own product to be represented.

Playstyles have already been removed as a result of 'Sudden Death' being removed, the 'Rule of 3', and 'Fixed Battlefields (Buildings,Objectives, etc)'. I don't think it's out of any variation of 'hating' aimed toward any particular path, but just listening to a group of players who would like to narrow the path to victory to something that is acceptable to them.

 LunarSol wrote:
Strength of Schedule remains the best, bad option, IMO. It's main weaknesses are that it feels completely arbitrary to players and heavily assumes that players will do their best even after they've lost. It's just the only system where I've consistently looked at the standings and felt they were a pretty solid listing of the results of the games I judged.
I'm not sure if you and I have had this discussion before or if it was someone else. Competitive 'Matched Play' events are trying to act like there are celebrity players, like the ITC rankings, without addressing the issue of not having a locked number of players. SoS works in established systems where every player/team would be playing every other player/team over the course of the year/season. SoS would work if there were a separate group of players showing up to multiple events where they only play each other. It fails when trying to implement that into an event where players are coming just for a handful of games.

I think this is why the 'Tale of X' gamers is actually exciting because the number of participants is restricted and everyone participating will be playing everyone else in the event. It's the best place to implement Strength of Schedule. It could be something worth looking into for actual 40k events. On the one side you'd have the top players in their own pool playing each other, which would provide excellent promotional media for GW, while allowing those who aren't in that group of players a chance to win their own local event. On the downside, we'd see even more copycat lists based on those players in the meta, and they'd never get to enjoy the biggest draw of major events, and that's playing someone outside your local meta.

-----
Since we're so close, the event is sadly going to be run like a 'Matched Play' event. We're just going to use the BRB missions, since they all don't have the GT20 booklet and there is not point in getting the GT21 booklet to use for 1 event and it doesn't matter.

They all like the idea of me tracking the event in a bit more detail to check the results and share them. I've got a few ideas floating around and I'll most likely share them here for discussion.

Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in gb
Longtime Dakkanaut




 Akar wrote:
Thanks for your Input @Lunarsol. I feel like you've shed some light on my frustrations and I'll attempt to respond.

 LunarSol wrote:
There's no good answer for tiebreakers because they're arbitrary by definition...
...Players often like VP based tiebreakers because they feel like they have control of their results.

This is HUGE part of why 'Matched Play' is the worst setup as far as scoring goes. VP's aren't being used as tie-breakers, they're straight up being used for ranking.


Wait...I'm not sure you're talking about the same as thing everyone else here. "Matched Play" is just a style of game with more formalised rules around mission/army selection and so on. In and of itself a Matched Play game doesn't tell you how to handle a linked series of games forming a tournament - the rules for a Matched Play game just tell you how to play a single standalone game. The GT mission packs also don't really tell you how to run a tournament, they just slightly alter the rulebook Matched Play missions and are theoretically refreshed every ear.

The traditional way of running a tournament is to have the final ranking done purely by number of wins, with tie-breakers only used after that. Usually the tie-breaker is total VPs or VP difference but other things can be used. That's what your OP implied was happening. If there's a tournament that's just using purely VPs to determine the winner/matchups I'd agree that's completely wrong but it's also extremely unusual to see that. If the tournament still uses wins to determine the winner and match-ups during the tournament then we're back to the question of how you sort people with the same W/L record. Again, I'd ask you if you have a better solution than either VPs or SoS? There are a great many TOs all over the world who have had to deal with this issue and I'm pretty sure practically none of them would agree either is a perfect solution but nobody has yet come up with a better one.
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

Slipspace wrote:
Wait...I'm not sure you're talking about the same as thing everyone else here. "Matched Play" is just a style of game with more formalised rules around mission/army selection and so on. In and of itself a Matched Play game doesn't tell you how to handle a linked series of games forming a tournament - the rules for a Matched Play game just tell you how to play a single standalone game. The GT mission packs also don't really tell you how to run a tournament, they just slightly alter the rulebook Matched Play missions and are theoretically refreshed every year.
So they're the same thing. Where are the instructions no how to run an event and how to pair players? Why is it not transparent for the players as well? Are the VP's tossed out after each round after ranking? Are they cumulative until the final round, then used to rank who the overall winner is?

Slipspace wrote:
The traditional way of running a tournament is to have the final ranking done purely by number of wins, with tie-breakers only used after that. Usually the tie-breaker is total VPs or VP difference but other things can be used. That's what your OP implied was happening. If there's a tournament that's just using purely VPs to determine the winner/matchups I'd agree that's completely wrong but it's also extremely unusual to see that. If the tournament still uses wins to determine the winner and match-ups during the tournament then we're back to the question of how you sort people with the same W/L record.
This is what everyone keeps trying to tell me, but it's simply not there. I'm not disagreeing with Wins/Losses being used, just that they don't seem to matter under the current system. Using VP's as a tie-breaker would imply that it's simply a +/- comparison at the end of the event. That's not what's being done. After each round the Winners are pooled and then ranked by their VP score. So the Wins aren't all treated equally like they should be and the problem was know before playing a game. What else would be the point of keeping track and reporting the actual VP's scored during a game?

When trying to make sense of any tournament results viewed there are too many questions. There is typically a W:L record followed by the scores of each round. While it's true that the top winners don't always have the highest overall VP total, the next bracket doesn't reflect using VP's as tie breakers for ranking the X:1 players. Quite a few of them highlight which round they lost, several of the winners ended up more favorably when they lost in an earlier round compared to those who lost in the later round. What's the explanation for that?

Slipspace wrote:
Again, I'd ask you if you have a better solution than either VPs or SoS? There are a great many TOs all over the world who have had to deal with this issue and I'm pretty sure practically none of them would agree either is a perfect solution but nobody has yet come up with a better one.
This thread is not about finding a better solution. It's about trying to understand how the stupid thing works. We already know it's possible to score more VP's and lose a game than someone who wins with less than that. Traditionally, after the first round, all of your 1-0 players go into pool and what? They're ranked by VP's? So the outcome is you get your 100 VP winners who had a favorable matchup now playing each other, while those with less favorable matchups are stuck playing each other. Two experienced players playing in the first round should have a similar result as two brand new players who are just starting. With no instructions, both appear to be penalized for their own games by the experienced player as both the experienced players. I just trying to make sense of this system one more time despite the frustration.

This message was edited 1 time. Last update was at 2021/07/05 12:59:34


Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in gb
[MOD]
Villanous Scum







Have you asked the TO what system they are using?
For what it is worth the two I could find quickly for 40k are;

Caledonian.
MATCH RESULTS SCORING
Victory Points Results & TP are automatically calculated based on Victory Point difference with a starting TP of 10 and +/-10 available

LADDER SORT ORDER
Total Tournament Points > Strength of Schedule > Victory Points > Random

STRENGTH OF SCHEDULE
Strength of Schedule (SOS) is calculated by:

Average of player's opponent's average TP per match (excluding byes)

Byes are excluded from the calculation so they will not increase or decrease SOS

PAIRING
Randomly among players on the same Tournament Points (excluding bonus points allocated for Time/Lists/Sports/Painting)

ITC;
In ITC format events, we suggest the following scoring guidelines:
1,000 Tournament Points + Battle Points for a win.
500 Tournament Points + Battle Points for a tie.
0 Tournament Points + Battle Points for a loss.
This format will determine pairings for you where the W/L/D record is primary. Players are paired in descending order, with the two highest scores playing one another, then the 3rd and 4th highest scores, etc., working your way down the rankings. Please note, the Best Coast Pairings app will automatically default to this scoring structure for you if you use it.



For Xwing and Legion they both use random draw first round, followed by Swiss pairings with Margin of Victory being the determinant.

On parle toujours mal quand on n'a rien à dire. 
   
Made in gb
Longtime Dakkanaut




OK, so the first thing to say is that the GT Mission Pack is just that - a mission pack. GW themselves don't actually detail a complete tournament structure, as far as I'm aware (I may well be wrong with this). This is in contrast to other companies who do. X-Wing, for example, has a fully detailed tournament structure that's followed by pretty much all tournaments worldwide. That said, there are a lot of 40k tournaments in the world that are run the same way so it's not difficult to find out how to run one. The pairing structure and how to determine the overall winner is pretty consistent across all of those tournaments.

Almost all 40k tournaments use a very simple system:

1. First rounds are paired randomly. Subsequent rounds pair people with the same W/L record together, often using total VPs (over all games so far) or VP difference (again, over all games so far) as the tie-breaker for people with identical records.
2. The overall winner is the person with the best W/L record. Players are sorted down the field using the same W/L record. Where records are the same, a tiebreaker is used. That's usually total VPs earned or total VP difference but could be strength of schedule or any other metric the TO decides on.

That's it. There is no "official" way to handle it but that approach is universal as far as I can tell. It's not perfect by any means, and it's at its worst in the earlier rounds because mismatches in the random pairings can take a while to even themselves out before they become representative of the relative strengths of the players. In your example, if two good players are paired in round 1, one will win and one will lose. the same is true of weaker players playing each other. However, over the course of the next round or two the sorting mechanisms of the system outlined above should end up with the weaker player towards the bottom as they start to lose games and the stronger player towards the top. Short of having everyone play everyone in some massive round-robin tournament you'll always have these issues to some extent, it's just the reality of running a tournament of 6-8 rounds with 50+ players.

I guess I'm just struggling to figure out what you're getting at with this thread. Nobody thinks the current structure is perfect but it seems to be good enough to give results reflective of the quality of players and armies at an event. The good players tend to consistently do well and poor players seem to consistently score poorly so it's clearly doing something right.
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

Slipspace wrote:
I guess I'm just struggling to figure out what you're getting at with this thread. Nobody thinks the current structure is perfect but it seems to be good enough to give results reflective of the quality of players and armies at an event. The good players tend to consistently do well and poor players seem to consistently score poorly so it's clearly doing something right.
The point of this thread is to make one last attempt to understand how the system functions. I have no idea where the competition lies for the reasons listed above and I disagree with it's existence. I'm not here to complain or vent about it, I'm genuinely trying to figure it out. Looking at the different events and the results shows inconsistencies on how they arrived at the result. I don't expect many systems to show a difference in the undefeated player. It's the following rankings that the problem shows itself. One tournament showed 2nd place losing round 1. Several others had 2nd and 3rd losing their games mid round. To me that shows that good players aren't consistently doing well since they don't necessarily get paired off against a more challenging opponent later rounds.

We're getting GW events this year, and there isn't a clear system in place on how to score. So far I've seen 3 different 'main' ways to score, including those listed above, and a few others that are a bit wild. Using VP's is a terrible implementation, but so is the Mission that is being played. I'm not even worried about the mission at this point and just trying to understand the scoring. If the scoring does manage to represent a competitive environment, then we can work on addressing the mission. That's all.

Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in gb
Longtime Dakkanaut




 Akar wrote:
The point of this thread is to make one last attempt to understand how the system functions.


I think this may be the crux of the problem. There is no single system for 40k. There are lots of similar but slightly different methods of sorting players at a tournament.

 Akar wrote:

I have no idea where the competition lies for the reasons listed above and I disagree with it's existence. I'm not here to complain or vent about it, I'm genuinely trying to figure it out. Looking at the different events and the results shows inconsistencies on how they arrived at the result. I don't expect many systems to show a difference in the undefeated player. It's the following rankings that the problem shows itself. One tournament showed 2nd place losing round 1. Several others had 2nd and 3rd losing their games mid round. To me that shows that good players aren't consistently doing well since they don't necessarily get paired off against a more challenging opponent later rounds.

We're getting GW events this year, and there isn't a clear system in place on how to score. So far I've seen 3 different 'main' ways to score, including those listed above, and a few others that are a bit wild. Using VP's is a terrible implementation, but so is the Mission that is being played. I'm not even worried about the mission at this point and just trying to understand the scoring. If the scoring does manage to represent a competitive environment, then we can work on addressing the mission. That's all.


There is no one singular way of doing this and any system will need to use some form of tie-breaker which is, by definition, arbitrary. I'm not sure why you're not understanding that. If you're having trouble figuring out how results are arrived at the tournament pack should describe the scoring system. The two most common ways to determine winner are:

1. W/L first, then VPs as the tie-breaker.
2. W/L first, then SoS as the tie-breaker.

Neither are perfect but they work well enough. There are some outliers. I believe some tournaments automatically give 2nd place to the loser of the final game, for example. Some tournaments are long enough to run enough rounds so only 1 person is undefeated but they'll still sort below that using one of the methods above, though sometimes they sort based on when a player lost rather than purely W/L record, with a later loss putting them higher up the standings. There is no perfect system for accurately ranking large fields of players after 5-6 games so anything that is used is imperfect.

The event pack for the official GW events details how the pairing and scoring system will work in a fair amount of detail. The pairings are done by matching you with someone with the same W/L record, but also someone who won and lost at the same stage as you did. So if you win in rounds 1 and 2 but lose in round 3 you'll play someone who did the same, rather than potentially being matched with someone who lost their first game. VPs are still used as a tie-breaker. That may be a fairer system but it's still subject to blind luck in initial pairings.
   
Made in us
Humming Great Unclean One of Nurgle






 Akar wrote:
Slipspace wrote:
I guess I'm just struggling to figure out what you're getting at with this thread. Nobody thinks the current structure is perfect but it seems to be good enough to give results reflective of the quality of players and armies at an event. The good players tend to consistently do well and poor players seem to consistently score poorly so it's clearly doing something right.
The point of this thread is to make one last attempt to understand how the system functions. I have no idea where the competition lies for the reasons listed above and I disagree with it's existence. I'm not here to complain or vent about it, I'm genuinely trying to figure it out. Looking at the different events and the results shows inconsistencies on how they arrived at the result. I don't expect many systems to show a difference in the undefeated player. It's the following rankings that the problem shows itself. One tournament showed 2nd place losing round 1. Several others had 2nd and 3rd losing their games mid round. To me that shows that good players aren't consistently doing well since they don't necessarily get paired off against a more challenging opponent later rounds.

We're getting GW events this year, and there isn't a clear system in place on how to score. So far I've seen 3 different 'main' ways to score, including those listed above, and a few others that are a bit wild. Using VP's is a terrible implementation, but so is the Mission that is being played. I'm not even worried about the mission at this point and just trying to understand the scoring. If the scoring does manage to represent a competitive environment, then we can work on addressing the mission. That's all.
Do you have a better approach, though? No one is really denying the current methods are flawed.

Road to Renown! It's like classic Path to Glory, but repaired, remastered, expanded! https://www.dakkadakka.com/dakkaforum/posts/list/778170.page

I chose an avatar I feel best represents the quality of my post history.

I try to view Warhammer as more of a toolbox with examples than fully complete games. 
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

quote=Slipspace 799296 11167486 null]I think this may be the crux of the problem. There is no single system for 40k. There are lots of similar but slightly different methods of sorting players at a tournament.

GW hosting 'Matched Play' events is proof that this isn't the case anymore. There is clearly a single system that is going to be used, which shouldn't be this hard to get an answer. If only there was some sort of publication that has been printed 3x that they could've used to clear this up. Or are you suggesting that each of the different events is going to use a different scoring system?

Slipspace wrote:
The two most common ways to determine winner are:

1. W/L first, then VPs as the tie-breaker.

1) If this one is in fact the one that's used, then winning has no real value. As I stated above, wins aren't treated with any sort of equal value since VP's are more important than winning. Sure Wins are still relevant, but they take a back seat to Winning, or else the mission wouldn't need to run for the full 5 turns regardless if there is only one player left. So after randomly matching up round 1, the subsequent rounds don't necessarily reflect a competitive environment since rankings are done based on how many VP's you receive. There is no difference in pairing rounds this way vs. keeping the random seating from round 1 to pair off players. This might as well be a single elimination style event. You're absolutely right that it doesn't affect who the undeafeated player will be, unless you end up in a event were you have multiple undefeated players. With the current mission, there is no available tie-breaker. You're using VP's to rank your players, not simply as a tie-breaker. You can see this failure in several of the events where players have lost in earlier rounds and ended up higher ranked than someone with the same W/L record who lost in a later round.

Slipspace wrote:
The two most common ways to determine winner are:

2. W/L first, then SoS as the tie-breaker.

2) This is pretty much the BP System that some TO's are falling back on. They might have there own reason for doing it, but for me it's because of the failure of the first system to more accurately reflect a competitive environment. The issue with this system is that it's incompatible with the current Matched Play Mission rules. This system tosses out VP's outside of determining who the winner is (as you say), and the game is effectively over. It does treat Wins equally which #1 doesn't do. It fails because tabling your opponent won't win you the game unless your already ahead in VP's. To be fair, it will allow a player who is behind in VP's to table their opponent and get that 1 VP over and still be counted the same as a player who scored 100 VP's. Without any secondary objectives to achieve from previous editions we still have no way to rank players for pairings and we end up in the same situation as #1. If we're going to back to this scoring system, then we need to restructure the mission back to something more functional.

 ingtaer wrote:

ITC;
In ITC format events, we suggest the following scoring guidelines:
1,000 Tournament Points + Battle Points for a win.
500 Tournament Points + Battle Points for a tie.
0 Tournament Points + Battle Points for a loss.

2.5) Just to cover all the bases, this systems is based off of #2 when introduced. All that's being done is assigning a points value to W/L/D for the simplicity of ranking players for pairings. The issue is still the same however since based off of this, the points for a win are the same regardless of round. A simple fix by increasing the points for later round wins, or consecutive wins. Without any BP modifiers from Secondaries available under the current mission the BP system fails.

-----
The tournament is a week away so without a set answer, we're going with the BRB mission using #1. I'll post the results here once I get it all together. Thanks for taking the time to respond.

Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in us
Longtime Dakkanaut




The old Battle Point system had every issue you're concerned about, and more. It feels a bit like a desire to go to a more archaic scoring and mission age, as you seem to be preferring outcomes that you say you dislike.
   
Made in us
Discriminating Deathmark Assassin




Out of my Mind

MVBrandt wrote:
The old Battle Point system had every issue you're concerned about, and more. It feels a bit like a desire to go to a more archaic scoring and mission age, as you seem to be preferring outcomes that you say you dislike.
This is about trying to understand the current system. I don't care if we never go back to the BP system, but there first needs to be a reason for replacing it. Which method will you be implementing at 'Matched Play' Tournaments? How does it address the issues?

This message was edited 1 time. Last update was at 2021/07/12 16:15:25


Current Armies
Waiting for 40k to come back in the next edition.

 
   
Made in us
Longtime Dakkanaut




 Akar wrote:
MVBrandt wrote:
The old Battle Point system had every issue you're concerned about, and more. It feels a bit like a desire to go to a more archaic scoring and mission age, as you seem to be preferring outcomes that you say you dislike.
This is about trying to understand the current system. I don't care if we never go back to the BP system, but there first needs to be a reason for replacing it. Which method will you be implementing at 'Matched Play' Tournaments? How does it address the issues?


At the US Open events, there are sufficient rounds to identify a single undefeated player for the "Top" Generalship award. Win Path Pairing (Which, in a randomly first round paired event, is the best available method for determining subsequent round pairings) is used to determine pairing and progression through the event. Competitive rankings are done by bracket, then SoS and VP per bracket, with Win Percentage being the first criteria naturally for each.

I'm still not sure any of your critiques make any sense, though I'm open minded as ever to people innovating solutions to problems. Is your main issue oriented around where to rank people within same win percentage?
   
Made in gb
Longtime Dakkanaut




 Akar wrote:
MVBrandt wrote:
The old Battle Point system had every issue you're concerned about, and more. It feels a bit like a desire to go to a more archaic scoring and mission age, as you seem to be preferring outcomes that you say you dislike.
This is about trying to understand the current system. I don't care if we never go back to the BP system, but there first needs to be a reason for replacing it. Which method will you be implementing at 'Matched Play' Tournaments? How does it address the issues?


That info is already in the GT pack and is the one I already described to you in a post you've replied to. I'm really not sure what your problem is any more. Initially it seemed like you just wanted to know how pairings were done but now it seems like you want to criticise the system used as not being fair or representative of player skill...I think. You've now been told what the official GW system is as well as how several other tournaments work out pairings but you still don't seem to be happy.

Again, I'd ask if you have a better solution?
   
Made in us
Humming Great Unclean One of Nurgle






TBH I don't really see what the big problem with wins<VPs is. The goal of the game is to win, winning is done by collecting the most VPs possible while denying them to the opponent as much as possible. It naturally follows that winning followed by VPs would be the method of determining rankings. No one ever claimed it was perfect.

Does it mean that blind luck of matchmaking can determine winners and loser over skill? Yes, it does. You know where else that can happen? Every fething match. Even going down to a single undefeated player is, to some extent, a product of luck. The roll for first turn has a greater impact on tournament rankings than any of the matchmaking processes discussed.>

Road to Renown! It's like classic Path to Glory, but repaired, remastered, expanded! https://www.dakkadakka.com/dakkaforum/posts/list/778170.page

I chose an avatar I feel best represents the quality of my post history.

I try to view Warhammer as more of a toolbox with examples than fully complete games. 
   
Made in us
Never Forget Isstvan!






There is literally no way to improve the current system.

You could implement an ELO system to rank players yes, but since every tournament is potentially ran or scored differently it would be pointless.

The closest thing to an overarching ranking system is the ITC. However it only manages to collect enough data over the course of a year. There simply is no way to play enough games of 40k for an ELO type matching system, and no real way to collaborate all events for it to be even worth it.

JOIN MY CRUSADE and gain 4000 RT points!
http://www.eternalcrusade.com/account/sign-up/?ref_code=EC-PLCIKYCABW8PG 
   
Made in gb
Longtime Dakkanaut




 NinthMusketeer wrote:
TBH I don't really see what the big problem with wins<VPs is. The goal of the game is to win, winning is done by collecting the most VPs possible while denying them to the opponent as much as possible. It naturally follows that winning followed by VPs would be the method of determining rankings. No one ever claimed it was perfect.

Does it mean that blind luck of matchmaking can determine winners and loser over skill? Yes, it does. You know where else that can happen? Every fething match. Even going down to a single undefeated player is, to some extent, a product of luck. The roll for first turn has a greater impact on tournament rankings than any of the matchmaking processes discussed.>


Exactly. Just looking back at a fairly small X-Wing tournament I played at the weekend it's obvious to see that match-ups and luck within matches will always have an impact in any system. We had one undefeated player, then another 6 people who had lost 1 game. The sorting for those was down to MoV (X-Wing's equivalent of VPs) and it's obviously an imperfect solution but it works well enough. A single bad dice roll at the wrong moment could conceivably have moved people up or down within that 1-loss bracket, or a different pairing at one point could have changed the order of opponents for other people and completely altered the tournament rankings.

There is simply no practical way to account for all of this. TOs know this and nobody claims their system is perfect.

One of the advantages of having VPs as tie-breakers is it's something each player has some control over. If I'm losing and don't think I can turn the game around at least I can try to maximise my VPs to mitigate the effects of a loss. It can also lead to problems if one person gets an easy draw early on and absolutely slaughters their opponent but - again - these are known issues with no easy answer.
   
Made in us
Fresh-Faced New User




warminster - United Kingdom

At FactoruM the BCP app is used for Warhammer 40,000 tournaments. I can highly recomment it.

FactoruM Games Venue is based at Unit 35-36 Deverill Road Trading Estate, Sutton Veny, Warminster BA12 7BZ (UK)
 
   
Made in gb
Executing Exarch





@ Slipspace, feel you on the x-wing scoring, played Ghost/Shuttle/Ahsoka for the tail end of 1.0 a list with a hella spikey MoV depending on if the Ghost survived and an X-1 or X-2 finish, a decent enough showing could land me anywhere from 2nd to 10th

And whilst no system is ever going to be prefect bearing scoring method in mind whilst making a list or at the table may help smooth out the rougher edges

"AND YET YOU ACT AS IF THERE IS SOME IDEAL ORDER IN THE WORLD, AS IF THERE IS SOME...SOME RIGHTNESS IN THE UNIVERSE BY WHICH IT MAY BE JUDGED." 
   
Made in us
Humming Great Unclean One of Nurgle






 FactoruM wrote:
At FactoruM the BCP app is used for Warhammer 40,000 tournaments. I can highly recomment it.
Worked SO well at the last LVO...

Road to Renown! It's like classic Path to Glory, but repaired, remastered, expanded! https://www.dakkadakka.com/dakkaforum/posts/list/778170.page

I chose an avatar I feel best represents the quality of my post history.

I try to view Warhammer as more of a toolbox with examples than fully complete games. 
   
Made in us
Never Forget Isstvan!






BCP works fine for any event with less than 300 players.

Once you get to 300+ players all trying to refresh their app at the same time for new round pairings the load on the server gets a bit rough.

The Guys at BCP could upgrade the servers, but since not all users are subscribers and im sure that costs a bit for the year, they are likely hesitant to do so.

JOIN MY CRUSADE and gain 4000 RT points!
http://www.eternalcrusade.com/account/sign-up/?ref_code=EC-PLCIKYCABW8PG 
   
 
Forum Index » Tournament and Local Gaming Discussion
Go to: