| | You are not authorized to post a reply. |
| | Author | Messages | |
Gnolaum
Sergeant
854 Posts
    
 | | 02/24/2006 6:20 PM |
| From my experience with Magic the Gathering (MTG) and D&D Miniatures (DDM) I have a few things to say about how the tiebreaker system works for D&D Miniatures.
It doesn't.
Now, why does it does not work. From my experience with MTG I can say it works fairly well. The defining feature that causes the current tiebreaker system to work for MTG and not DDM is games per match.
In MTG each match of the tournament consists of 3 games. A person may win 2-0, 2-1, or lose 1-2, 0-2, and this is recorded and tracked in DCI reporter, and used in every tiebreaker.
In DDM each match of the tournament consists of 1 game. A person may either win or lose. There is only two possibilities win/loss. As such the precision caught by the tiebreaker system is extremely diminished (half over 1 round, quartered after 2, etc, geometric progression in loss of accuracy).
Take a simple tournament of 8 players. After the first round, on average, there will be 4 different levels in MTG, and only 2 in DDM. After the second round there will quite often be no ties in MTG, but after the second round in DDM there will be 4 ties (one for each of 1st, second, third and forth).
The key problem is that the tiebreakers attempt to capture some information about how well a participant did in a match, and succeed in MTG and various other games that have multiple games per match, but fail almost completely to do so in games like DDM that have only one game per match.
I see it happen in almost every tournament that I run were at least 2 people are exactly tied across the board in points and all tiebreakers. I believe that the system then resorts to entry time into the tournament for a final tiebreaker. Often this tiebreaker is used to determine who gets into finals and who does not. Not fun, and not fair.
Even when the tiebreakers actualy break ties, they are of a dubious nature for DDM.
So the key problem is that the system attempts to capture information regarding how well a player did in a match, but fails to do so. How might it succeed in doing so? The answer to this question is easy. Incorporate victory points.
I would sugest the following be used as tiebreakers. They may seem rather complicated, but they are really actual almost identical to what is currently being used: instead of reporting a match as 2-1 as in MTG simply record the VPs (207-156), the ratios/math might be a little bit harder, but that is done by a computer anyway. Implementation of this in the DCI software would be trivial. It should already be able to do it.
1) Opponents Match Point Ratio
An average of (2) of all you opponents.
2) You Match Point Ratio average.
If you win game 1 207-156, then your ratio for this match is: 207/(207+156)= .57
If you win game 2 207-198, then your ratio for this match is: .51
You ratio for the tournament thus far would be: .54
3) Since in DDM there is no difference between match and game, the translation would result in the exact same as the first tiebreaker. I would propose simply dropping the third tiebreaker. Since the ratios provided by (1) and (2) are much more precise, I doubt there would ever be a need for a second tiebreaker, let along a second
4) Entry time into the tournament. I believe this is the hidden tiebreaker that the system currently uses.
I believe this system is a much more accurate method of breaking ties for games such as DDM that have a point system, and only 1 game per match. | |
My online store http://store.hoardsters.com
Use quick search to see scans of every stat card! | |
| 
Wrackspawn
ChristopherGroves
Warlord
6093 Posts
| |
Vrecknidj
Warlord
10445 Posts
United States
| | 02/24/2006 8:33 PM |
| If we could get everyone to keep track of points, and to do it correctly, this method might be useful.
I'm no fan of unjust procedures, and I don't think that the current system is the best it can be. (Why have a match worth 3 or 0 when there are no other current options? I think if the system is binary, then 1 and 0 pretty much cover it.)
By going by opponent's strength, a few problems are introduced (like people dropping out, people getting a bye, etc.). I'd like to see a better system all around.
But, I don't think we'll see it anytime soon.
Dave | |
Knowledge Arcana editor issues 5-9, Phoenix Lore Magazine editor, assistant editor for Rite Publishing;
My Trade Thread and My Reference Thread; Winner of WBC IV, IX and XIII; Rule #0: bshugg is always right! | |
| Knight of Argenis
Corim Danex
Warlord
6809 Posts
West Valley City, Utah
| | 02/24/2006 8:46 PM |
| | Byes are also a strange thing to consider. | |
"Look to God and live." Alma 37:47
Vindicated Champ of Hippogriff (Arcadian Hippogriff) and Uncommon Horse | |
|
Tactician
Sergeant
888 Posts
| |
chattan
Sergeant
393 Posts
| | 02/25/2006 3:14 PM |
| quote:
Originally posted by Corim Danex
Byes are also a strange thing to consider.
As we noted in our recent UT Open. Corim had a bye the first round. I defeated one of our top players, Drakkengi. We each won our 2nd round game. We met in the 3rd round and each of us were surprised to hear that Corim was ahead of me in the standings. His bye counted for more than my win. That ain't right.
Of course he then beat me and ended all talk of tie breakers. [B)]
chattan | |
| |
|
| | You are not authorized to post a reply. |
|
| |
ActiveForums 3.7 | You must be signed in to participate in the
games. |