"Wild Bill's Rumblings Of War"- A Tournament

[JonS]

my head hurts

[Peter Svensson]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Treeburst155:

Hmmm... what about ties? I'll have to think about that. It does happen.

<HR></BLOCKQUOTE>

I'm not sure how NABLA handles this, but I suggest that if two people are tied for, for example, second place, they split the points for second and third place. If second place is worth 10 points and third 9, then each of the players would receive 9.5 points. The player with the next lower score would receive the points for fourth place, i.e. 8 points.

If three people were tied for second, they could split 10+9+8 points three ways, and so on.

[Holien]

God this has taken me awhile to get my mind round what the difference will mean.

:eek:

I think it will be a good thing as it will reward people who lose well.

So if I lost a game (bound to happen) and got 20 points to the other players 40 I would have just got 20 points in the old system.

But the new system means I would get effectively 100 points (aka 11pts new system)if I was the best loser of that group.

Yes this makes sense as I get rewarded for losing well compared to the rest of the group. It still does not mean that I will get into the finals for losing but will give me a fighting chance if I happen to draw the wrong end of the sticky stick in each scenario.

Yes lets do it.

H

P.s So will AAR's be worth 10% aka 1.1 points?

[Treeburst155]

Peter,

Good idea for handling ties!

Holien,

Yes, AARs will be worth 1.1 point maximum keeping it at 10% of the top score possible for a scenario. The AAR points will be added to players' score at the END of the tourney.

Jon,

Trust us, this system is much better.

Treeburst155 out.

[ 09-25-2001: Message edited by: Treeburst155 ]

[JonS]

Oh, I agree its better. No problem there. Just getting a tad confused trying to work out all the permutations in my head...

[Treeburst155]

Look at it this way, Jon. Suppose a scenario is way out of balance and you get wiped out 93-7. Chances are, those who also have to play this scenario from the same side as you will also do poorly. Your 7 points might actually get you a high "final" score for that scenario when it's compared to the other poor guys who had to play that scenario from that side.

Now look at the reverse. Your opponent in your section gets 93 points to your 7 points. When you compare that 93 with all the other players who played that scenario from that side his 93 score may be very average. You could actually get more points for a scenario than your opponent even though you lost the game! This would be a somewhat extremem situation, but it is possible if a scenario is way out of balance.

Treeburst155

[JonS]

yeah, yeah - I understand that part of it just fine It's the number of games, sides, scenarios and matches, and what all that meant for the tourney that I was getting fouled up with.

I'm going to pretend I'm a starfish now. Starfish don't have these problems in their lives.

[hands_over_ears][chanting] I'm a starfish, I'm a starfish, I'm a starfish[/chanting][/hands_over_ears]

There. I feel better already

[John Kettler]

Treeburst155, WineCape and troops,

The proposed scoring system (or variant thereof) seems fine to me, enables frighteningly realistic scenario development, and addresses some knotty statistical issues as well arising from the relatively small number of trials used to establish play balance. If Wild Bill and crew got ten in per scenario, the minimum for any statistical validity, I'd be amazed. That's not a crack, merely a recognition of limited resources, many scenarios to create, tweak and test, and very little time.

Essentially, the proposed scoring system is a form of what in my school days was termed "grading on the curve" and was sometimes used for tests and other difficult matters. Grades were assigned based on relative performance against one's classmates. Of course, the person who did really well on such a test was none too popular with those who did poorly, since it set the bar pretty high in consequence.

Regards,

John Kettler

[Buckeye]

This new scoring system sounds like a very good idea, and could be extended to many other tournaments. As John says, we didn't play-test every scenario 10 times, not only because of limited time, but also because you would really need 10 tests conducted by players blind to the scenario, not 10 by folks who have already played the scenario. So we can never guarantee a scenario is balanced. Until now, the only way to deal with this problem was to have mirrored games, but that doesn't work nealry as well when the forces are fixed from the start, because then the attacker knows what forces the defender has, and vice versa.

I'll be very interested to see how this scoring system works. Mike, I assume WB and the rest of his team will have access to the standings?

On another front, I'm pleased everyone is enjoying WB's scenarios. I've read two of the AARs so far, and they were both entertaining. They mirrored my experiences in some ways, but were different in others. I look forward to reading more.

[tabpub]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Treeburst155:

Look at it this way, Jon. Suppose a scenario is way out of balance and you get wiped out 93-7. Chances are, those who also have to play this scenario from the same side as you will also do poorly. Your 7 points might actually get you a high "final" score for that scenario when it's compared to the other poor guys who had to play that scenario from that side.

Now look at the reverse. Your opponent in your section gets 93 points to your 7 points. When you compare that 93 with all the other players who played that scenario from that side his 93 score may be very average. You could actually get more points for a scenario than your opponent even though you lost the game! This would be a somewhat extremem situation, but it is possible if a scenario is way out of balance.

Treeburst155<HR></BLOCKQUOTE>

But then look at a less extreme situation:

Say that in a certain scenario the average score for the 28 games for one side is ..um

56 say. Without being precise, let us say that 27 of the scores fall in the range 50-60 and 1 score was 85. In the abovementioned system of scoring, the player with an 85 would score 11 pts and the next highest(say it was 60) would score 10. Is that necessarily equitable? The one player that was able to "shine" above the rest would not reap the full benefits of this accomplishment.And the last place person would be shut out because he was a 50.

Think this over. Take that average score for a side in a specific scenario. Compare each players score to that average. If you are above, you get that many positive points;below, and you get that many negative points. At the end, add each players "comparative" scores together to get a grand total.

Now this system, if used, could come back to bite me in the ass, as I might be the poor sod that is "below average" all the time, but it really does reward someone who can "make the impossible, possible".

I now retire back to lurker-dom. Thank you for your time.

[Treeburst155]

Tabpub,

Right off the top of my head I would have to say, "You have a very good idea there!"; but I have to think about it a bit first. How many points to give for AARs is one thing to think about since points awarded for scenarios will have a variable maximum. I don't want AARs to sway the standings too much. I just want them to make a difference between players who are running fairly close to each other.

More important is the fact that the guy who scores 92 points from a side of a scenario, and scores 25 points more than #2 for that side/scenario, may have played a very weak opponent. He would be rewarded for his fortunate matchup with the weak player. Who is to say whether it was his skill or his opponent's lack of skill that brought about the high score? Staying with the original "Nabla" system would tend to make the tourney a tighter race, I think, as extreme scores one way or the other would be tempered. It will be more difficult for someone to run away with the tourney, and more difficult to fall hopelessly behind. This might make the race more exciting.

EDIT: Buckeye, you can view the standings by clicking on the "Standings" link in my sig. There's nothing there yet for the Wild Bill tourney other than an empty chart however.

Treeburst155 out.

[ 09-25-2001: Message edited by: Treeburst155 ]

[Treeburst155]

OK, how about this scoring system. First the average score is determined for each side/scenario. Points would be assigned to players based on this average in the following manner:

+/- 4 of average=0 points

+5 to 9 of average=1 point

+10 to 14= 2 points

+15 to 19= 3 points

etc....

If you are more than 4 points below the average for that side/scenario your points go into negative numbers in the same fasion. This system rewards overwhelming victories and punishes crushing defeats to a certain extent while avoiding punishing the lower scores in a situation where the scores are all very near the average. What think ye, gentlemen?

Treeburst155 out.

[ 09-25-2001: Message edited by: Treeburst155 ]

[tabpub]

That would work also. Kinda like golf. Establish the par(average)and then a set point for being so far over. That would work also and probably be simpler.

Along that vein, then grade the AAR's like school A=4 B=3 etc. then a good AAR with a bad result could balance things out a bit.

Watch out though, you guys could be reading novels over there for a month....

[Treeburst155]

I don't really want to get into the business of grading AARS for several reasons. I think 2 extra tourney points for a good AAR and 1 point for a mediocre AAR is plenty of incentive to write. I figure the winner will probably score a total of 28-30 regular points + 14 bonus points for 7 AARs. If the winner scores 44 points total and AARs are worth 14 points (for 7 AARs)I think that's plenty of incentive to write.

Treeburst155 out.

[Nabla]

Hello everyone!

I started a new thread for discussions about the new tournament scoring system.

The Balanced and Unbalanced Battles Tournament Scoring

[JonS]

Wild Bill and the Scenario Design Team,

One observation for future scenarios that will require AARs. Could you include lots more named features on the maps, so we can reference them. I'm having a little trouble describing the action just in terms of N-S-W-E locations. Some of the maps are fine, but others could use a little more naming.

As a minimum I would suggest naming each victory location. Other useful spotnames would be villages, crossroads and main roads or linear features (rivers, treelines), and of course major hills and forested areas. Also any isolated buildings/farms. Hmm, that's quite a list - but only some will be relevant to any given map, and it certainly would make the job of writing it up clearer and easier

Thanks heaps

Jon

[Treeburst155]

We've got 47 games passed out with 37 to go. Two of the 47 have already been completed. This is really great progress! Thanks guys!

Now back to the scoring situation. Although we've only heard opinions from a few on this thread, they have all been in favor of the "Nabla" scoring system. I'm so convinced it is a much better way to score this tourney that I'm going to implement it even though I haven't heard from most of the participants regarding it. They will find out soon enough.

I'm going to use my compromise solution between Tabpub's suggestion and the original Nabla system. For details on how this works read the last couple pages of this thread carefully. If you still have questions just email me and I'll try to clear things up for you.

There will be a web page available to view the results of individual games. This will require some time to get going. I may be collaborating with Peter Svensson on this. My original standings page for this tourney is now useless. All game results will be posted to this thread so check here from time to time to see who beat who and by how much.

If anyone has any serious objections to the new scoring system I need to hear from you ASAP so I can explain to you how and why you are wrong.

Now, get back to your foxholes!

Treeburst155 out.

[JonS]

TB155,

giving teh double blind nature of this tourney you might want to consider not revealing which scen the players of the completed games were fighting. Its not much info, but knowing how others went in the same scen could affect the play of participants who have yet to complete (or start) that scen - especially in light of the new scoring system.

Regards

Jon

[Treeburst155]

Jon,

If I did what you suggest there would be no reason to even bother with a statistics page. I could just report the scores of completed games on this thread without revealing sides or scenarios, just names. Players would almost be completely in the dark as to how they stand.

This is fine with me, and saves me lots of work (the web page); but people may not want to be that much in the dark for the duration of the tournament. I kinda like the idea though. Wild Bill and the testers, WineCape,

and I will know what is going on and get to enjoy the race.

I guess we need to take another survey. How much do you guys want to know about games that are completed? Do you want to know which scenario two guys played? Do you want to know who played which side? Do you even want to know the final score of others'games at all?

Keep in mind that without specific information about a completed game (players, sides,and score) you will have no way to compare your performance with others during the tournament. There can be no page showing the ranking of players since that can't be determined until all games are completed.

I could post win/loss statistics but they would be of dubious value with this scoring system. So speak up, guys. How much info do you want to know when it comes to completed games?

Treeburst155 out.

[JonS]

But with the new scoring system we won't know anyway, until all the games for that scen are finished since its based on relative - not absolute - performance. At least, I think that's how it goes :confused:

[Treeburst155]

If a webpage listed all the scenarios twice in columns, and all player results were posted to the proper column (Allied-Kommerscheidt for example), and ranked by score; players could get a fairly good feel for how they are doing compared to others as the tourney progresses. I'm for keeping you guys totally in the dark now that I've been thinking about it. Nobody knows anything except their own game results until the very end when all will be revealed. I like it!!

Treeburst155 out.

[Olduvai]

Greater minds than I are heavily engaged in this scoring discussion, and I bow to their wisdom. I am happy with any scoring system.

[Spanish Bombs]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Olduvai:

I bow to their wisdom. I am happy with any scoring system.<HR></BLOCKQUOTE>

Ditto on that. Thanks for everything again.

But where are my damn reinforcements!!!

[Bertram]

Regarding the scoring system:

I liked the first option best, for various reasons.

First it is simpler, something I am in favor of .

Second, I have the feeling that the system that gives more points for scores diverging more from the average, though rewarding brilliant play, also puts more emphasis on one time results and unbalanced games. Though I have not made any real statistical analyses, this is based on the following:

Suppose a game is balanced, and the opponents are also (and nothing freaky occures in the battle). Scores will vary closely around 50-50. So most people will get one point plus or minus.

Enter various variants. One opponent really goofes (gooves?) it (he had a row with his girlfriend, or just a string of bad luck). The score is 30-70. They get plus or minus 4 points for this one, and the other players need a 'strange result'to, to be ever able to catch up.

Enter one real bad player. All win against him. As a result the score against him will be on the edge of the bell curve for that scenario. This means that your score against him wil be 4 points, and playing against him will be important for winning the tournament.

And finally some scenario's lend themself more to varieing scores then others. Winning a scenario where the score is (on both sides) between 0-100 (lets say a tank shootout, with a limited number of tanks, where you can get lucky, or have a string of bad luck) will be much more important then winning a hard fought battle where everyone scores between 45-55, even if you manage a brilliant 40-60 win.

If I remember correct this is also the reason why bridge uses the order of points to give scores, and don't use the actual points scored in the carious games.

Bertram

[ 09-27-2001: Message edited by: Bertram ]

[Enoch]

Much like designing voting systems to determine group preferences, designing a scoring system to determine the winner of a tournament is fraught with problems.

Any system that is proposed will have some drawbacks and some advantages. The question is what do you want to maximize/minimize? If you want to minimize the effect of outliers then you might use the bridge system as suggested. However, this has the effect of 'punishing' players whose scores' group together. For example, if 5 players score 50, 51, 52, 53, and 54 then the player scoring 50 is harshly penalized if you simply rank these scores.

The modified golf system that Treeburst is comtemplating using minimizes the problem of closely grouped scores and slightly rewards outliers.

There is no real right answer to this question. Treeburst should simply decide (with our input) what we want the scoring system to accomplish. I personally think the modified golf system is a good compromise between the normal system and the bridge system.

(edited bekause i kant spel)

[ 09-27-2001: Message edited by: Enoch ]