A CM Tournament idea that accommodates unbalanced attack / defence scenarios

[noob]

If eight Allied players played the same attack / defence scenario against eight Axis players in a CM knockout tournament, eight Allied and Axis scores would be generated, so if the top four scoring players from each side qualified for the next round, the CM scenario used for the qualifying round would not have to be balanced.

[JonS]

Nabla, meet Noob. Noob, meet Nabla.

[noob]

Nabla, meet Noob. Noob, meet Nabla.

Too wordy

[JonS]

BTW, Noob, your proposal in the first post in this thread is possibly flawed in that it sems to assume that the top four players on side A are all "better" than the bottom four players of side B. At the limit B8 could be a better player than A1, but under your system A1, A2, A3, and A4 will go through, while B5, B6, B7, and B8 all get relegated.

[JonS]

Too wordy

It's complex, that's why it's wordy.

It's fair though. And works well, as long as you have a tourny director that has at least a basic understanding of how it works ... and thank god TB155 did, because I'm not sure anyone else truly understood what was going on

Edit: but note that the system described in that link was based on the CMx1 scoring system, in which all scores were always in the range 0-100. Some work would need to be done to accomodate CMx2s more flexible scoring arrangements.

[noob]

BTW, Noob, your proposal in the first post in this thread is possibly flawed in that it sems to assume that the top four players on side A are all "better" than the bottom four players of side B. At the limit B8 could be a better player than A1, but under your system A1, A2, A3, and A4 will go through, while B5, B6, B7, and B8 all get relegated.

No, it doesn't assume anything about the bottom four side B players, it states categorically that the top four players on side A are better than the bottom four players on side A, and likewise for side B.

I don't see what the problem is, as long as a player who loses a CM battle better than his comrades can qualify to the next round the scenario can be unbalanced.

[noob]

It's complex, that's why it's wordy.

It's fair though. And works well, as long as you have a tourny director that has at least a basic understanding of how it works ... and thank god TB155 did, because I'm not sure anyone else truly understood what was going on

I still don't see why it has to be complicated though, as long as there is a reasonable tie breaker system its pretty straightforward the way i see it, the top halves of each side qualify.

[JonS]

No, it doesn't assume anything about the bottom four side B players, it states categorically that the top four players on side A are better than the bottom four players on side A, and likewise for side B.

I don't see what the problem is, as long as a player who loses a CM battle better than his comrades can qualify to the next round the scenario can be unbalanced.

What if all 8 B-side players win the battle? It could be because the battle is unbalanced, but it could also be because all the B-side players are just better than any of the A-side players. But under your proposal 4 x A and 4 x B would go through, regardless of their relative quality.

In other words, you seem to be assuming that the A and B pools are evenly balanced. That might not be a valid assumption. (And this is exactly why Nabla ended up being so complex, because it cross-leveled between groups. It didn't care how balanced the scenarios were, nor how balanced the pools were in terms of skill level. It teased out the best players, even if they were playing on the same side of the same pool)

[Wreck]

Noob, your system is what Nabla would generate if using only one game per tournament round. Nabla is a generalization of what you are doing so as to compare results in a set of games and pick a set of best players.

Your system has the advantage of nearly trivial computability. It has the disadvantage that using a single game to determine superior players is subject to a fair amount of luck.

[noob]

Noob, your system is what Nabla would generate if using only one game per tournament round. Nabla is a generalization of what you are doing so as to compare results in a set of games and pick a set of best players.

Your system has the advantage of nearly trivial computability. It has the disadvantage that using a single game to determine superior players is subject to a fair amount of luck.

I'd take the luck to have the simplicity

[noob]

What if all 8 B-side players win the battle? It could be because the battle is unbalanced, but it could also be because all the B-side players are just better than any of the A-side players. But under your proposal 4 x A and 4 x B would go through, regardless of their relative quality.

In other words, you seem to be assuming that the A and B pools are evenly balanced. That might not be a valid assumption. (And this is exactly why Nabla ended up being so complex, because it cross-leveled between groups. It didn't care how balanced the scenarios were, nor how balanced the pools were in terms of skill level. It teased out the best players, even if they were playing on the same side of the same pool)

I see what you are saying, so i would suggest, for the sake of simplicity, that we ignore the complexities and just pretend that my suggestion is a valid way of determining the best player

[noob]

In other words, you seem to be assuming that the A and B pools are evenly balanced. That might not be a valid assumption.

You are correct in pointing out it may not be a valid assumption, however one could err on the side that it was a valid assumption given that, if any random number of players were divided into two pools, there would probably be more of a balance of skills in each pool than an imbalance between the pools.

[JonS]

Sure, maybe. But random is a tricky concept.

For example, say you call a tournament, with 16 places available on a first come first served basis. And the players are allocated spaces as they express interest. Even leaving aside the global nature of this forum causing issues, the pool of players is self selected in a non-random order.

[noob]

Sure, maybe. But random is a tricky concept.

For example, say you call a tournament, with 16 places available on a first come first served basis. And the players are allocated spaces as they express interest. Even leaving aside the global nature of this forum causing issues, the pool of players is self selected in a non-random order.

But the pool of players could be re ordered in a random way.

[JonS]

Sure, but you still have what is basically the same problem. Assigning them randomly cannot assure you that you've got four 'good' and four 'bad' players in each pool.

OTOH, doing it that way is a whole heck of a lot more tractable than trying to set up a fully Nabla-ised tourney

[womble]

If all you're trying to determine is an overall winner, it doesn't much matter, so long as the single "best" player is selected at each stage. In the scheme Noob is suggesting, there's an assumption that the eventual winner will be in either the top 4 of A or the top 4 of B, which is probably a fair assumption. Nabla will be able to produce eventual more accurate rankings for all the players, but until the final, Noob's method works to balance out scenario imbalances.

[slysniper]

The Method is fine for what you are wanting to do. Yes, JonS has a point, but who cares. The point being, you are after the best player, no single round elimination tournament can gareentee that. JonS is a Jerk most of the time, just loves to be a pain and gets into always having another view point.

But back to what you are suggesting. So who cares if allied player number 5 does not move on even though he was better than the 3rd and 4th german player who does get to move on. That is the luck of the draw, and he still did worse than 4 other Allied players, so he is likely not the best player, so who cares if he is better than some others that move on. You are truly after only rewarding the Best Player when it comes down to it.

That is not even going to happen because it all will be random as to who has the best scores, much will depend on who you are up against. So those with the best scores might be great players or they just played poor players from the other side.

I used a similar format in my out of the dust tournaments, but they are not single elimination. They were more of a round Robin concept where everyone remains in. But with time and games you can rate players and see the true better players work to the top of the Leader Board, you might want to see what I did in them threads.

But even in them I would see a inferior player up in the list better than where they should be. But given enough games, that type of system is the best, where some type of rating is given to players.

The good part about it was I could use any scenario for the tournaments, Balence is not important, so you could test the players skills in many types of Missions.

Actually I was thinking of running a "Out of the dust Tournament #4"

but I have not mastered or had times to generate scenarios in the new system. I would need donated scenarios that have never been released or made available donated to the Tournament. I was thinking two matches per round.

[Tarquelne]

Sure, but you still have what is basically the same problem. Assigning them randomly cannot assure you that you've got four 'good' and four 'bad' players in each pool.

The obvious answer is to hold a Nabla-tourney to determine the "good" and "bad" players, and then assign them accordingly for a Noob-tourney.

This gives you the complexity of Nabla at the front but adds a dash of uncertainty for the final winner, essentially combining the worst of each system. But it solves the assignment problem.

pretend that my suggestion is a valid way of determining the best player

Just call it how you're going to determine the tournament's winner and you don't need to pretend anything.

[dieseltaylor]

I was writing at WeBoB on this type of set-up for a Tournament as I think CMBN makes scenario balancing very hard. So bite the bullet and make it very simple like a Knockout with a Plate.

The way I view it you end up with an Allied and German winner - and if it were desired those two have a final battle and the Plate competition run simultaneously will provide a par score and the player who exceeds the par by most wins overall.

The way scenarios are constructed needs to be given some thought in terms of trying to more finely graduate the scores. The WeBoB database shows a very high number of extreme scores recorded - about 50% or more are 80/20 or better/worse.

My pet theory is that the loony provision of invisible walls around small maps means that artillery is made stronger as units bounce of the walls and run back into fire, vehicles cannot leave the map, and units simply do not retire even when all is lost. I am not quite sure what aspect of realism the invisible walls represent but scenario designers are really going to have to appreciate this aspect of design needs to be corrected from the default.

However an WEBoB annual tourney is on the cards and its the results from the early tourneys that has provided the data which is giving pause for thought about how it is constructed.

I am a great fan of NABLA but given the huge variability in spotting times for tanks [and possibly for other troops] it would seem to be giving too much precision to games where generally good players should succeed but luck will have a big role.

[A Canadian Cat]

So noob, interesting discussion you started up here The more I read the more I think a NABLA scoring system is the way to go even though it is difficult to grok by most players.

But the bottom line is if you want to run a tournament then just set the rules and be clear up front. If you followed the Farm tournament that Bimmer has been running you can see that there are many happy forum members who had fun playing. I for one would do it again even though I was eliminated in the first round by one of the players that is going deep. Would some kind of round robin or NABLA scoring system have been better? Who really cares if everyone has fun and the rules were clear at the start.