I don't like the idea of bringing Gambit into the mix for rankings.

Not that I have anything better to suggest :/

I used to play chess, and there is a system of points for wins, losses and draws... but here's the fun thing, you get less points if a high-ranked player beats lower-ranked opponents, while lower-ranked opponents get more points for winning against the higher-ranked guy (and doesn't lose as many points when you lose to the higher ranked guy).

Therefore, to be #1, you have to take on tougher and tougher matches until you eclipse #1.

Just a thought.

I like the idea of starting around 1000 pts. (1138 points would be silly, but very iconic!) I don't think the amt of points won by should make a difference, honestly. I think that, let's say, you beat someone who is at 1100 rating, and you are at the flat 1000. A percentage of the difference in ratings should be how much you go up, say, 10%, so you'd go to 1010. Perhaps 10% for a loss to a lower ranked player, so the loser would go to 1090. Then, for losing to a higher ranked player, you would lose 5%, and someone who beats a lower ranked player gains 5%. This is a system that rewards for beating higher players, and only slightly for defeating lower ranked players. Okay, just realized that, if we're all at the same level, this won't work mathematically, ha ha. So, how about a minimum of 5 pts gained for wins!

Did I explain my craziness well enough?

Okay, another crazy idea just came to me! This is a goofy one, but I like it. Everyone starts with 1000 points. How much you gain or lose depends totally on you and your opponent! You see your opponent's squad, which is a total hose for you. You bet 5 pts, your opponent agrees. Whoever wins gains that much, whoever loses has that much taken from them. But they want to wage much more, you settle with an average, no games ever go over 20 pts! This is a goofy idea, but, hey, it is an idea!

Someone shoot me down or agree, at least I'm thinking....ouch, that is what that sharp pain is!!!

That's how the chess system works, kinda.

Because (to me at least) it's irrelevant to whether you take half your opponents' pieces, all of them, or just a few pawns in your win.

A win is a win, and factors (killpoints, gambit) toward that win are irrelevant.

Well, the DCI algorithm is pretty simple. It uses a base system of 16, which is why you start at 1600pts. You basically take each participants current rating, insert them into an algorithm, and it gives you an odds of victory percentage. Then that percentage is multiplied by 16 to get the score you go up for a win, and down for a loss. It works well because if you play a guy with the same rating, its an 8/8 split. If its like Dean playing a kid with a 1600, its a 2/14. So Dean wins, and he gets 2pts. The kid loses 2pts only. Its pretty simple to plug into a computer program I think.

So not saying we should use their system, but I don't think it will help us to try an reinvent the wheel.

Well I don't mean to shoot down your ideas Fool, sorry lol. I just don't think that game rules/luck/squad strength should have an impact on your rank. IMO rank should depend on the rank/record of the players involved.

One of the things I thought of - and I'm just throwing this out there - is simply dividing the losers rank by the winners rank and increasing/decreasing points as per the result. So for example, somebody with a rank of 70 defeats somebody with a rank of 100, the winner goes up 1.4 points, and the loser goes down 1.4 points.

I think factoring Gambit can only make it more reliable. Anyone can kill a few pieces and let time run out. But to actually engage and kill those big pieces...In a lot of games that are timed, it can get tiresome to just see the opponent hang back once he has a 5 point lead.

Well I certainly think that Gambit should be a part of the game, but I don't think Gambit should factor into your rank as a player...

I too don't think the score of a game should have any bearing on how it affects your ranking. What you guys are forgetting is that so much of those factors are not in your control. I should know, I play fast, but I don'y always finish my games, do you think that is my fault? Matt and I finished our 6th round Masters game in 14 minutes, to completion, he had 150 points by that time.

I'm not sure if it's possible -
but is there a way to "Cross refrence"
sort of like how interest can be recalculated on a mortgage if you make frequent payments

Looking into the DCI reports I see I've played WAY more games than some of the top 100- yet I'm ranked in the lower 200

I'd like something more reflective, not just wins/losses - but perhaps frequency?

mayby I'm asking too much-
not every town has 3 diffrent shops that can play at
or a store where all the workers can sanction an event on the spot...

In keeping with the above, I do not think Gambit or Specific game elements should be a factor-
I DO Think there should be a seperate ranking for each build level
like there is for Limited vs Constructed
BTW I'm # 75 in Limited

So here's a bit of a brainstorm that (as far as I know) isn't included in DCI. In addition to earning points for winning games at a tournament, why not some sort of bonus for winning an entire tournament? This could be a universal bonus applied to any tournament anywhere (perhaps scaled to the number of players--more bonus points for winning a 16 player than a 4 player tourney).

Or, the other way to do it would be to reserve the tournament victory bonus for a small number of "sanctioned" tournaments. If the goal is to value "the best playing the best", part of how you could make the "Qualifier" or "Championship" tournaments matter is to announce ahead of time that "the winner of the Ultimate Vassal Tournament of Infinite Justice will receive 25 bonus rating points," or "the champion of the SWM Gamers Garden State Classic, held this year in scenic Elizabeth, NJ, will receive 15 bonus rating points," or whatever. That way, the ratings reflect that a small number of prestigious tournaments have a luster that goes beyond beating the guy who only collects OT characters down at your LGS.

I have to agree with Billiv15 if I understand him correctly. We maybe don't need to reinvent the wheel. It seems to me the dci ratings system works fine it's the op structure that is screwed up. If we used Vassal & had reporting hubs in each state or region (or other areas of the world) that were all reporting then problem solved?

I'm cool with just sticking to DCI rankings so long as Vassal is a competitive playing field. I'd be worried about legal shenanigans from WotC, but I'm sure somebody else here could offer better insight into that than I.

I think you need to keep using DCI but forget about WotC OP department organizing stuff for you, if you are serious about continuing to support the game.

The more I think about this, the more difficult it is for me to go along with what's happening. I mean I'm concerned and I'm glad you guys are taking up the charge, but I'm still not on board with divorcing WotC OP and dividing the community to the point that there aren't enough people in either group to really support anything.

If they get to keep making all the money, and they barely care about as doing OP for us now as it is, then imagine how ecstatic they're going to be when they get all of the profit with NONE of the responsibility...

Say goodbye to retail and league kits if you do this, guys.

Finally, I'm certainly not going to support them as a customer any longer if someone else is handling all of the actual leg work of organization. I can still play with the stuff I have and trade for pieces from people who were going to buy product anyway.

I would hope we would take one last crack at it over the next year before going ahead. I know Rob and Chris are on our side, the trick is going to be getting through to the higher-ups.

I would love to chat with Sarah for a half an hour or so about this, but I am willing to go over her head if I think I can reason with someone else. I don't know though, I still remember having a long talk with Rachel Kirkwood a couple of years ago about how the game is so unique heavy, which meaning have to really go after boosters for rares, and now we have some kick-ass commons and uncommons. I don't know if that had anything to do with my chat, but I'd like to think it did.

It gives me confidence about persuading them on this issue, too.

I posted this in the other thread as well, because I hadn't realized yet that there was a specific thread just for the points system.

I'm brainstorming here, but what about something that was based on the overall win/loss ratios of the players? If you are 10-0, you should be ranked near the top, whereas, if you are 0-10, you should be near the bottom. Now, obviously, if you're 10-0 against scrub players, you shouldn't be ranked as high as someone who is, say 9-1 against other top-ranked players. So I think it would need to be a mix of the two.

Something like, the amount of points you gain for a win is dependent on how tough your opponent is (how high they are ranked). Obviously, you should earn more points for defeating a higher ranked player (regardless of their relation to your own ranking). Similarly, you should get less points for defeating someone who is far below you. In this way, people who consistently beat the same people all the time, and quickly rise to the top of their little group, will see diminshing returns on their ranking.

Then, your overall ranking is somehow increased or decreased by an some given amount, say, 1000 points, but multiplied by your winning percentage first. So, if you've won 80% of your games, you get 800 points. Or something along those lines.

It would be tricky, and I haven't done any math yet to see if it could be broken, but then you'd only be getting a set amount of points for being undefeated (not something that would continually increase with more wins). Could also do it as a percent increase of your ranking (say you have 1200 points just from games, then you're 80% w/L means you get an increase of 8% 96 point). Or something like that.

Again....brainstorming.

And I also like Boris ideas for getting in more tough with Sarah and Rachel.

Your suggestion (20 points per win, -5 per loss) seems the same. Except it eliminates the best part of ELO, which is that beating up on newbs or lousy players repeatedly nets you less and less points.



This called ELO ranking and is exactly the system that DCI already uses.

The problem with DCI specifically is that SWM does not include enough "cross polination" of players to keep the ranking in line. Nothing short of required "title defenses" of some kind can really achieve that which falls outside of the scoring system itself.

ELO is a fundamentally sound system which is easy to impliment and for which the abuses and weaknesses are clearly known. It's as good as any base scoring system you can choose. From there it's all about how you handle the details.

I would still like to use the ELO system. Its a great relief to check my scores some times and see that if (IF) I were to beat Dean in a rated game, then my score you jump up! but If I were to loose, then it wouldn't fall so much.

I also like the idea of Titles and title defense. Winning the Yearly championship, qualifiers or any other major tourney through the year (so long as said tourney is accessible to ALL) should give your ranking a boost. But if you can't hold on the title.. it should drop a little. As well if you continue to hold it year to year get some point boost as well.

All on top of the normal points gained/lost for playing.

We seem to have two issues here. First, whether this is a replacement for, or in addition to, the DCI system.

Second, how and what to include in a ranking system.

I’m not going to debate the merits of DCI. It is what we have, and what we will continue to need to keep prize support coming.

So, here is my take on the second part:

Rather than determining a number system, why don’t we take a step back and determine the requirements for the tracking system. If we know WHAT we want it to represent, developing the system will be much easier.

Here are some requirements to discuss. They either came from the discussion above, or my own thoughts:

1) Easy to input data/measurements
2) Measurements need to be deterministic (wins/losses) rather than subjective (squad strength).
3) System needs to assess relative strength of competitors.
4) System needs to include some measure of ‘scale of victory’ (easy win, or difficult)
5) The system needs to normalize the battle results by the point value of the contest.
6) Established weighting factors for each of the measurements.
7) System includes frequency of play modifier for ranking.


Other reqirements???