Public Demand: Who wants to see Titans of TH, RD, BoB clashing?

[Fionn]

Mike,

"I understand what you are saying now. This fact is what makes the Integer Nabla highly competitive. Beating somebody 95-5 will not generate a big lead. You would still have to play very will in the other games. The big victory offers no padding. Note that the converse is also true. A big loss will not knock you out of the running."

No offence but I don't think you do understand me now. I am NOT interested in trying to ensure that someone who loses BADLY is still in with a shout.

As far as I'm concerned the % scores SHOULD be used since they offer more internal discrimination than NABLA integer scoring ( which is, in effect, 100%, 75%, 50%, 25% ). Again, I'm not interested in keeping people in the running if they lose BADLY a few times. If they lose really badly then they probably SHOULD be out of the running IMO.

My basic issue is what the fact that the two types of games are not comparable (IMO) and, as such, we should settle on one game type OR have everyone play both game types against every other person in the tourney.

If I lose 5 to 95 then I should be SOL as far as I'm concerned. ( another reason I prefer not to use the NABLA integer score BUT as I have said before this is a minor issue compared to the fundamental flaws inherent in the 4 scenarios/3QBs issue and I think that should be ironed out first).

Also, as I've said I amn't trying to shove anyone into some weird system they don't like. I am quite happy to simply step aside if everyone wants to use what I feel is a flawed system. No bother to me at all.

[Treeburst155]

Redwolf,

Feel free to contribute. You are a scoring expert. Your opinions on issues such as being discussed here carry lots of weight with me.

If we go mirrored QBs I could always set some rather harsh restrictions on towed guns.

Treeburst155 out.

[Wreck]

redwolf:

We have limits. I have signed up for a tourney thinking 5 games (6 players, 1 game each), and now it is 7. That's fine. Mirrored battles require 14 games. That's too much.

As for scenarios: we have 5 for ROWII. I don't see why people think scenarios are incomparable with QB-style games. But if people really think that (and I strongly disagree), then another solution is to get two more scenarios.

As for me, I think that there are MANY aspects to what makes a truly titanic CM player. One is to be able to dominate in QB MEs, as is seen at TH. Another is dominate in QB attacks. Yet another is to do well in scenarios, that is, with forces you don't pick yourself and interesting terrain. And one more, even rarer skill, is to do well in a fog of war about the scenario itself, where you don't even know if it is balanced. This is what I learned in ROW and Nordic Wannabees. I think it was valuable.

So that's four things I think we should emphasize in Titans. My proposed games, and scoring system, reflect that, along with the time limits we all have.

I have not seen a detailed proposal from anyone else, beyond "7 QBs". But even with that: what about scenarios? What about imbalanced games? If you don't think these are important parts of CM mastery, well, say so. Let's be upfront about our biases.

As for the whining issue: if we play just 7 games and declare a winner, yes, someone will probably whine. I won't, though. (Well maybe a little about that lucky gun-damage...) If the winner can't take a little whining in stride, then he's not much of a winner. And if the non-winners can't get over their loss, even if it was unfair in retrospect, I am sorry for them. Ultimately, it is still a game and I hope we can all keep that in mind.

[Treeburst155]

Originally posted by Fionn:

Mike,

Maybe they can't ALL be balanced but IF 1 more person was added one could:

a) ensure everyone played 2 defences as allies, 2 as German, 2 attacks as German and 2 defences as German..

This may be true, but players would still meet "the best" in a different situation than others would. Also, are you sure this can work out? That's a lot of parameters for the scheduling. The schedule would have to be figured out manually. It might be quite a puzzle. I'd have to play with it awhile to see how difficult it is.

________________________________________________

In order to keep Fionn interested, and to level the playing field to the extreme why don't we just go with QBs. Each player choosing ALL the parameters for one scenario as I mentioned earlier. To this we will apply the Integer Nabla scoring. Why? Because NO scenario is balanced IMO. Simply tallying points is the worst way to score. This applies even if we manage to equalize attack/defend/sides because players will attack and defend and play different sides in different scenarios.

Treeburst155 out.

[Redwolf]

Wreck, I meant to say: if a scheme is assumed to require doubling the number of games it should be chosen.

The point I was going to make in the rest of the posting is: if we want to reflect different conditions, different types of CMBO games, then there are two ways.

The first way is mixing QBs and scenarios.

The second way is doing seven similar games under conditions that have elements of all play styles. Premade maps with own force selection in mixed weather/time/ground conditions and mixed attack/defense/ME seem to have many aspects of all play styles for me. Maybe we can improve it by half-selected forces, you get a frame force and can choose like 1/3 of the points.

I think it will be difficult to find a compromise over the issue of truly unbalanced games. Fionn is right that they distort things from the majority of normal CM games played in the past and that they can screw one player even in the Nabla system. And you (Wreck) are right that they are an important aspect of true mastership. It looks like a classical "both are right" argument to me.

Last but not least it seems to me the curves in the Nabla system may be worth modifying. This tournement has players who know what they do, will not surrender and should generally be of comparable strenth. The modifiers that apply for the effect of very high wins and very bad losses may be worth modifying here.

Just my usual unhumble opinions.

[Treeburst155]

Fionn,

Out of curiosity, why should the high average score win, especially when that average could have been attained by playing the weakest player in the scenario that is the most unbalanced? Keep in mind, I'm going on the assumption that ALL scenarios are unbalanced.

Treeburst155 out.

[Treeburst155]

Redwolf,

Modifying the curve for this tourney would be a very good idea IF we were using the full Nabla system. We are using the playoff system that does not require an accurate median (because we won't have one).

How is a player screwed by an unbalanced scenario? The Nabla system is specifically designed to address that issue. An unbalanced game is simply another test of CM skills.

Treeburst155 out.

[ June 10, 2002, 04:14 PM: Message edited by: Treeburst155 ]

[Redwolf]

Originally posted by Treeburst155:

How is a player screwed by an unbalanced scenario? The Nabla system is specifically designed to address that issue.

In this tournament you can expect the player with the advantage to press it through. There is few chance for the other player to look exceptionally good even if he is very good.

That would all be fine if everybody was playing everybody in every scenario.

But here you play everybody just once. Let as for the sake of the argument assume that Fionn and Swamp are the best two. Fionn has to attack Swamp in a scenario that is unbalanced in favour of the defender. Then Fionn may be waxed no matter how good he is, with comparably little difference between good or bad play on Fionn side.

This robs Fionn of the opportunity to score important points against Swamp, since this is the only scenario he plays Swamp in. He also don't have the chance to beat the here-assumed-to-be only competitiotor (Swamp) for victory in a scenario where he can press an advantage through.

You can now start playing with the Nabla curve modifiers for high loss/win. You would have to emphasize the difference between losses like 15% and 10% to reward Fionn for playing well in this scenario. But if you do that you badly screw people who drop into these loss percentages in balanced scenarios.

An unbalanced game is simply another test of CM skills.

That's true, but it is arguable that it was a very rare mode of play for past CMBO battles. If personally think it's great.

[Treeburst155]

The fundamental assumption behind Fionn's arguments, I think, is that balanced scenarios exist. This I disagree with. Only by mirroring battles can you achieve true balance.

This means some form of comparison between players who play the same side in a battle must take place to determine relative skill if mirrored battles are not used.

Why not split into two groups of four. Each player plays a mirrored battle against three others (six games). We would then have a mirrored QB playoff between the section winners.

The inequity here is simply the fact that one section may be stronger than another section. All else is as fair as can be. We can even use raw CM point totals.

Treeburst155 out.

[ June 10, 2002, 04:40 PM: Message edited by: Treeburst155 ]

[Broken]

Fionn,

If the four Scenarios were scored under Nabla and the three QB were scored by points, you are correct that the situation would be unfair. But what Wreck and I are saying is to use Nabla in all SEVEN games. I know that in an earlier part of this thread the dual scoring system was mentioned, and maybe that is the source of confusion.

Also, Fionn, you are right that opponent mismatches will add some degree of imbalance to the tournament if you only play each player once, from one side of one of the seven Games. This is inherent in any system where we don't play every opponent from both sides of every Game. But, as you pointed out earlier, it would be nice to keep our social lives.

The only practical way to be completely fair would be to have only one QB type where we play every other player from both sides. This would still be 14 games apiece and sounds pretty boring, to me anyway.

As for non-linearities, well, hey, most things are non-linear. We can't solve this problem out to third order, or we will be here forever.

So, I like the seven game format, even with it's inherent unfairness. We get to play each other once and each game is different. If I get the toughest players in games that favor their strengths, so what? Life isn't fair, but it can be fun. The winner gets bragging rights, but if he didn't win by much, he shouldn't brag too hard.

I think the integer Nabla system has too much "quantization noise" as shown by the Fionn and Wreck examples above with the 90-79-79 vs 80-80-80 scores. Why not just scale to 100 the four scores playing one side in a game? This would reward the player who, on average, proportionately scored the highest.

For example, if player A and B had the best Axis scores in four of the games, with scores of 80-80-30-30 for A, and 60-60-50-50 for B, these scores would be scaled to 100-100-60-60 for A and 75-75-100-100 for B. B wins even though their unscaled averages were the same. This is fair since B had proportionately higher scores than A. In the first two games, A outpointed B by 33%. In the last two games, B outpointed A by 66%.

So, here's a proposal:

Seven games in QB format with maps that have picked for their interesting terrain and maybe "spiced up" a bit.

Each player plays every other player once and is sided with every player three times. This so nobody gets sided with, say, Swamp six games out of seven.

All four scores on one side of a given game would be scaled to 100, (e.g., 50-40-30-20 would scale to 100-80-60-40). Highest total score over seven games wins. If people don't like this system, I can live with Integer Nabla as long as it is applied to all games.

Sizes could range from say 1500 pts to 3000 pts (4500 for the attacker).

A mix of ME, Allied attack and Axis attack.

Each game would have the same purchase rules (e.g., P76) for all players, but they could differ from game to game. I like the proposal (TB155's?) that each player submit one set of playing conditions. Maybe the playing conditions include the map type, battle type (ME, etc), and purchase points as well.

Purchase could be blind, pre-look, or a mix. I think pre-look is only necessary for really out-of-the-ordinary maps (see the RD Tourney Round 4 AAR thread, the map had a large body of water running through it).

[Deathdealer]

This is getting little bit out of hand - actually this is getting mostly ridiculous... I somebody really wants to be nominated BEST CM-PLAYER EVER IN THE WHOLE ***KING UNIVERSE be my guest. I don't mind

Back to point - I think Fionn has more beans for me though I like other arguments too . It seems to me that to get really even we should play 7 scenarios AND 7 QBs so that everybody plays everybody once in BOTH categories but I think that would be awful burden to bear. I don't think I want to go on with this tourney forever as I'd like to play others occasionally too. Sooooooo, one way or the other - let's play as I'm getting bored here

[Redwolf]

Originally posted by Treeburst155:

The fundamental assumption behind Fionn's arguments is that balanced scenarios exist.

But slight inbalances will be winnable by the disfavoured person. Even the Titans cannot be sure to ruthlessly push the advantage through.

The Nabla system will take care of the slight inbalances nicely.

Why not split into two groups of four. Each player plays a mirrored battle against three others (six games). We would then have a mirrored QB playoff between the section winners.

It doesn't seem to me the disagreement is bad enough to bail out yet.

[Treeburst155]

Yes, this is out of hand, but it's very interesting IMO.

Redwolf,

This is the same argument Fionn is using. It is somewhat foggy to me. I just can't quite grasp it. If true, then the same would apply to all scenarios since all are out of balance IMO. The only good answer is mirrored battles. Split into two sections of four, the number of games will not get out of hand.

Broken,

Your scoring proposal is interesting. I must think about it. You may be onto something there.

Treeburst155 out.

[Treeburst155]

I see what you are saying, Redwolf. The assumption then is that the Titans are so good that they need to play scenarios that are as balanced as possible if there is no mirroring. This because they will most likely be able to exploit any significant advantage in a disproportionate manner,regardless of their opponent's skill, thereby screwing the unfortunate opponent who plays the significantly weaker side. This logic is really hard for me to grasp. It seems "fuzzy" to me. I'm just not sure it's true I guess. Like Wreck says, even if it is true, is it a big enough issue to overcome the luck inherent in any CM game?

Treeburst155 out.

[ June 10, 2002, 05:17 PM: Message edited by: Treeburst155 ]

[Redwolf]

Originally posted by Treeburst155:

I see what you are saying, Redwolf. The assumption then is that the Titans are so good that they need to play scenarios that are as balanced as possible if there is no mirroring. This because they will most likely be able to exploit any significant advantage in a disproportionate manner,regardless of their opponent's skill, thereby screwing the unfortunate opponent who plays the significantly weaker side. This logic is really hard for me to grasp. It seems "fuzzy" to me.

The argument must be fuzzy, no doubt. It requires assuptions about how these players will score, and they are hard to make.

I feel that the Titans will drive the "unlucky" opponent much deeper into the loss zone, and the Nabla system doesn't work as good in deep loss zones as for medium losses. Because deep in the losses you get into the same regions that get special treatment to prevent plain unlucky people from being punished too bad.

The other argument is that you only play each person once and if we see one given pair clash in an unbalanced scenario we won't get the real data.

Don't understand me wrong, I think the ability to use unbalanced scenario in tournament is the best since sliced bread. But I think it is more fun with more uneven opponents, in bigger tournaments, and where you don't care about this specific opponent that much.

I'm just not sure it's true I guess. Like Wreck says, even if it is true, is it a big enough issue to overcome the luck inherent in any CM game?

The question here is how unbalanced we talk.

!!!SPOILER - don;t read if you plan to play the Nordic scenarios!!!

To take examples from the Nordic tournament, I think "The Aftermath" is not suitable for this tournament because it is basically poker, a kinda of poker CMBO player were not trained for in the last two years and hence it doesn't say much about the Titans of the CMBO area. But "Economy of Force" would. Both were inbalanced, but Economy of Force had the inbalance not in number of units or in the terrain, but in details like lower unit quality and surprising placement. That is what the CMBO expoert is supposed to be able to handle.

Maybe a compromise can be found along these lines.

[Fionn]

Mike

"Out of curiosity, why should the high average score win, especially when that average could have been attained by playing the weakest player in the scenario that is the most unbalanced? Keep in mind, I'm going on the assumption that ALL scenarios are unbalanced."

Well, I am assuming that we'll end up defaulting to 7 QBs. With QBs there are no initially unbalanced scenarios so the weakest player is likely to get his ass handed to him by everyone ( and not just by the guys who get him in really unbalanced scenarios).

Here's my proposal ( kinda similar to some that have gone before and I figure I should put my money where my mouth is so others get the chance to tell me how **** my system is ).

1. 7 QBs. Players get to see the map before playing ( to ensure that no-one gets totally screwed by a weird map). Of course if the map is properly screened by whoever is running the tourney and creating the maps then pre-view wouldn't be absolutely necessary since it would be that guy's job to make sure the maps are fair). All maps would be auto-generated using the same settings ( e.g. light woods, modest hills, rural.... I personally find that gives nice good maps.. Not too much cover to allow massive infantry rushes but still enough cover to maneuvre tanks etc effectively.) Purchase rules are Panther/76 with no exceptions.

2. Players get assigned nationality and role at random ( the organiser would be responsible for making sure that a fair enough ratio of defence and allied is given... e.g. it would be unfair to have one guy be the Allied defender in 6 of his 7 games ).

3. players play ALL the QBs.

4. The scores of all QBs are added together and the top 2 players go into a play-off situation in which they play a mirrored scenario against eachother. Highest combined score over those 2 PBEMs wins.

That way some weird statistical quirk can't completely determine the winner. Also the 1st and 2nd have to duke it out and the community would get a GREAT AAR out of it.

Mike,

I think the key point re: balance ( which is NEVER perfect) is that the initial balance in an equal points ME or a 1.5:1 ratio attack/defence is much closer than the balance in most scenarios. Since all Titans players are going to be absolutely brilliant at capitalising on advantages ANY minor initial advantage will be turned into a major game-winning advantage very quickly.

We played a game once remember? Ok, think of that map as being a bit biased in my favour ok... You never made it out of no-man's land IIRC. FWIW though I played that map in testing against 4 other people and none of them made it more than 200 metres forward either. My point is that when I gave that map to others one attacker actually managed to win so not all players can capitalise on advantages to the same extent we can.

I've sat in on games top players have had with others and it can, sometimes, just be a really horrendous massacre. I've sat in on these games and I can tell you that it overcomes luck. It is like watching a sheep being led to the slaughter., Their opponents never even have a chance. Not even the slightest chance. Just remember our game and remember that that was a mild example. That was only a very mild example.

[Treeburst155]

I've come to the conclusion that "The Aftermath" was highly dependent on luck. Note the Allied scores below. They swing wildly from big wins to big losses. One section is missing as they could not be scored properly. My reasoning could be flawed here, but with scores spreading clear across the spectrum for one side I suspect a high luck factor.

Allied Scores For The Aftermath

Mr Johnson........76

Strider...........72

Cpl Carrot........70

Cogust............69

Wreck.............65

Thumpre...........64

Tero.............62

Jack Trap........61

Tuomas...........59

Kingfish.........58

ciks.............57

Juha A...........54

a1steaks.........53

Lopaka...........52

_____________________

PasiN............39

Fate.............38

Romeoman.........37

Ugbash...........35

J. Porta.........29

vskalex..........20

dragoon..........19

[Treeburst155]

Fionn,

I believe the scenario we played was a no-win for the Allies even in a Titan vs Titan situation. I will grant you that. Even a Titan attacker would not do much better than an average attacker in that one as long as there is a Titan defender. You have illustrated your point quite nicely by bringing up that scenario.

I'm finally convinced that a contest among Titan's should include only scenarios that are as balanced as possible. However, and you knew there would be a "however", simply totalling raw scores is definitely inferior to any form of the Nabla system if we don't play mirrored games. This because ALL scenarios are unbalanced.

Treeburst155 out.

[Fionn]

Mike,

Cool.

QBs with summation of total scores or balanced scenarios using NABLA are AOK. I'd prefer QBs but scenarios are cool by me.

Thanks mate.

[Treeburst155]

Assuming we go seven QBs there are still some balance problems. The attack/defend ratios and sides will not work out. This, IMO, does not matter as long as the Nabla System is used. Players will always be compared with players who were presented with an identical initial situation. If a guy ends up playing five games as Allies and he feels he's better with German forces there is nothing we can do about it. Any inequities left can only be dealt with through mirrored games, which nobody seems to want to do.

I think that each player should be able to set the parameters for one scenario. One unlucky guy will see his scenario parameters thrown out, but there's nothing we can do about that.

If all can agree on the 7 QB plan with Integer Nabla scoring (the only type we can use)we can move on to actual selection of scenario parameters and force purchase rulesets.

Everyone simply needs to post their scenario parameters here. I would suggest actually stepping through the QB screens to do this.

I'm ready to move on at this point. Let's have some scenario parameters. Be thorough. Don't omit any item where there is a choice.

Treeburst155 out.

[Broken]

Treeburst,

What you said, man. QBs with Nabla scoring are fine by me. I like a percentile Nabla better, but integer is OK, too.

[Treeburst155]

Broken,

I was just going back to look at your scoring proposition right now. I just rushed through it before. This stuff is interesting to me, which is why I like running tourneys. I'll have some feedback for your scoring variation soon.

Treeburst155 out.

[Treeburst155]

Broken,

I think your idea is an improvement over the Integer Nabla IF I understand it correctly.

Given four Allied scores for one scenario:

69, 54, 38, 36

I understand the scores would be adjusted as follows:

69 becomes 100, 54 becomes 78.26 (because 54 is 78.26% of 69), 38 becomes 55.07, and 36 becomes 52.17

IOW, the high score becomes "the whole" and the other scores are raised to the same percentage of the whole they had with the original high score.

I think this is a good little tweak to The Nabla System. I like it!

Contested/unclaimed VL points will still be split based on a player's percentage of the total points scored. 60-30 would yield a score of of 66.67-33.33 for example. From here we would move on to the normalization to 100.

I also like Fionn's idea of a mirrored playoff between the top two considering the unavoidable inequities of uneven comparison and attack/defend/sides distribution. Remember though, the scheduling program WILL optimize these things as much as is possible with priority going to an equal number of comparisons with all other players.

Treeburst155 out.

[Broken]

Treeburst,

You understood me correctly. Thanks for looking into this.

[Fionn]

"Contested/unclaimed VL points will still be split based on a player's percentage of the total points scored. 60-30 would yield a score of of 66.67-33.33 for example. From here we would move on to the normalization to 100."

Sounds good to me.

Mike,

Can you summarise where we are at now with respect to rules etc.