rec.games.trading-cards.jyhad

the rating system is flawed

25 messages from 12 participants · 16 September 2004 – 18 September 2004
original thread on Google Groups

Stefan Ferenci

as much as i like the new rating system, i still think it is very much flawed. i played in the ne qualifier this year and finished 3rd with 1 gw and 5.5 vps. 40 player participated in the qualifier i recieved 98 points. at gencon i played the final qualifier with 90 players participating. i finished 5 with 2 gw and 8 vps but only recieved 79 points. although i had a much better showing at a larger tournament i recieve 19 points less. my point: i think the size of a tourney is not adequately represented in the new ranking system. stefan

Stone

"Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> a écrit dans le message de news: 41497430$0$10840$3b21...@usenet.univie.ac.at... > > although i had a much better showing at a larger tournament i recieve 19 > points less. > > my point: i think the size of a tourney is not adequately represented in > the new ranking system. you didn't have a better showing, since you were rank 5 in the larger tournament, compared to rank 3. The rating system focuses on player standing, which of course is in turn determined by GW/VP. In short, standing is more important than GW/VP if you want to get more rating points-the size of the tournament is not really the issue here, IMO. Stone

LSJ

"Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message news:41497430$0$10840$3b21...@usenet.univie.ac.at... > although i had a much better showing at a larger tournament i recieve 19 > points less. A 3rd place finish is a better showing than a 5th place finish. -- LSJ (vte...@white-wolf.com) V:TES Net.Rep for White Wolf, Inc. V:TES homepage: http://www.white-wolf.com/vtes/ Though effective, appear to be ineffective -- Sun Tzu

Daneel

On Thu, 16 Sep 2004 08:24:09 -0400, LSJ <vte...@white-wolf.com> wrote: > "Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message > news:41497430$0$10840$3b21...@usenet.univie.ac.at... >> although i had a much better showing at a larger tournament i recieve 19 >> points less. > > A 3rd place finish is a better showing than a 5th place finish. My cents: Stefan wrote: > i played in the ne qualifier this year and finished 3rd with 1 gw and > 5.5 vps. 40 player > participated in the qualifier i recieved 98 points. > at gencon i played the final qualifier with 90 players participating. > i finished 5 with 2 gw and 8 vps but only recieved 79 points. NE qualifier: 3 out of 40 with 1 GWs and 5.5 VPs. Objectively he was better than 37 participants. Relatively he scored in the upper 7.5%. GenCon: 5 out of 90 players with 2 GWs and 8 VPs. Objectively he was better than 85 participants. Relatively he scored in the upper 5.5%. That should probably mean that he should gain more Rank Points for the GenCon performance. "Besting" 85 players seems harder than "besting" 37. Accumulating 2/8 seems harder than accumulating 1/5.5. I think that in general the closer you get to one end of the distribution of the participants, the more indicative the distance from the other end will become (as opposed to how humans frequently view it). Scoring 5th place in the preliminaries is very different from scoring sixth, yet in a tournament with 90 players the objective performance is likely almost identical - "besting" 85 or 84 people is a very small difference. Same with the low end: scoring last is in for a "lemon" prize, but being 89th or 90th is quite an insignificant difference. Most representative of performance is perhaps the actual GW/VP scored, regardless of any position. Or the number of players who were "bested". In my opinion, of course. ;) Bye, Daneel

Joshua Duffin

"Daneel" <dan...@eposta.hu> wrote in message news:opsefcx3...@news.chello.hu... > On Thu, 16 Sep 2004 08:24:09 -0400, LSJ <vte...@white-wolf.com> wrote: > > > "Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message > > news:41497430$0$10840$3b21...@usenet.univie.ac.at... > >> although i had a much better showing at a larger tournament i recieve 19 > >> points less. > > > > A 3rd place finish is a better showing than a 5th place finish. > NE qualifier: 3 out of 40 with 1 GWs and 5.5 VPs. Objectively he was > better than 37 participants. Relatively he scored in the upper 7.5%. > > GenCon: 5 out of 90 players with 2 GWs and 8 VPs. Objectively he was > better than 85 participants. Relatively he scored in the upper 5.5%. > > That should probably mean that he should gain more Rank Points for the > GenCon performance. "Besting" 85 players seems harder than "besting" 37. > Accumulating 2/8 seems harder than accumulating 1/5.5. > > I think that in general the closer you get to one end of the distribution > of the participants, the more indicative the distance from the other end > will become (as opposed to how humans frequently view it). Scoring 5th > place in the preliminaries is very different from scoring sixth, yet in a > tournament with 90 players the objective performance is likely almost > identical - "besting" 85 or 84 people is a very small difference. Same > with the low end: scoring last is in for a "lemon" prize, but being 89th > or 90th is quite an insignificant difference. This is all true, sure. Before the current system went into place, I argued for a scoring method that ran a little more linearly (or on a curve, really) from 1st place down to 50th (or whatever last happens to actually be), but what we ended up with is a system that weights making the finals much more highly than anything else. Everyone here has read the scoring system page, right? :-) http://www.white-wolf.com/vtes/index.php?line=veknRating.html The bonus points for placing between 1st and 5th in a tournament are very significant to overall points earned - especially in larger tournaments, because the size-coefficient on points earned applies to the finalist bonus but not to the standardized "points for VPs, points for GWs" earned in all rounds. For example, you get 10 points for 5th (times the size coefficient) but 30 for 3rd (again, times the size coefficient) and 90 for 1st (still multiplied by the size coefficient). That's why 3rd is normally going to score better than 5th, even if the 5th is at a significantly larger tournament than the 3rd was. I don't think this system is significantly flawed relative to many other ways it could be done. I do, as you say, think that it would make more sense to slope the points earned down past the finalists more, but given that tournaments have always been played with a "finals are the most important game" system, it does make some sense for the finalists to earn substantially more points than non-finalists. Josh when a black cat crosses my path

Stefan Ferenci

LSJ wrote: > "Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message > news:41497430$0$10840$3b21...@usenet.univie.ac.at... > >>although i had a much better showing at a larger tournament i recieve 19 >>points less. > > > A 3rd place finish is a better showing than a 5th place finish. > not compared to the size of the event (see daneels post) a more extrem example e.g. ben peal won an 8 player tourney in boston (smash and grab) in 2003. in four games he scored 2 gw and 7 vps. he recieved 98 points for that win. thats more points than i got for going into the finals at a ninty player tourney while i scored 1 more vp in one less game (these are just examples). all i am saying is that size should matter, and the size coefficient is clearly not efficent enough. stefan

Curevei

>That should probably mean that he should gain more Rank Points for the >GenCon performance. "Besting" 85 players seems harder than "besting" 37. >Accumulating 2/8 seems harder than accumulating 1/5.5. As placement already factors GWs and VPs, it's not necessary to double count them. The current system only minimally double counts them, well, compared to what you get for making the finals, at least. If you don't make the finals, it's your entire score. You can't accurately compare absolute numbers between tournaments which is why the previous system was so ludicrous. Could, for instance, have two 40 player tournaments where one sees no time outs and lots of sweeps because the players' skills are so uneven and the other sees lots of time outs and table splits because the players' skills are so close. Actually, it doesn't matter why you see one or the other, but these are not unreasonable reasons as to why you might see this difference. Placement can't take into account the total skill of all players in the event, but it fully takes into account, to the extent that any rating system that will be used is likely to do so, relative skill for the event. (Which is kind of why I'd rather see placement be the primary scoring method for everyone.) Now, getting back to a possible flaw, it could be that 3rd shouldn't count any more than 5th as players don't think that much in terms of coming in 3rd. Going to depend upon prize allocations, but sometimes players won't value 2nd any more than 5th. My sense is that people, looking from the outside, mostly care about 1) did the person make the final table 2) did the person win. And, sure, somewhat getting back to the thing about beating more players, you do get kind of screwed if you don't make the final table. Stefan's result may be the problem of valuing 3rd and 5th so differently, but whoever came in 6th in either event probably deserves a lot more since they both beat large numbers of players where being 6th in a 15 player tournament (apparently the standard for the point system) is not nearly as impressive. I don't understand why pure placement couldn't be used. But, recalling the hassle that was instituting the latest fix and the mindnumbing uselessness of the previous system, I could see not worrying that much about this problem.

LSJ

"Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message news:4149b565$0$8024$3b21...@usenet.univie.ac.at... > LSJ wrote: > > "Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message > > news:41497430$0$10840$3b21...@usenet.univie.ac.at... > > > >>although i had a much better showing at a larger tournament i recieve 19 > >>points less. > > > > > > A 3rd place finish is a better showing than a 5th place finish. > > > not compared to the size of the event (see daneels post) Yes, it is, by virtue of what it means to be a better finish. At the very least, it is open to interpretation and one of those interpretations is that 3rd out of 40 is better than 5th out of 90. > a more extrem example > e.g. ben peal won an 8 player tourney in boston (smash and grab) in > 2003. in four games he scored 2 gw and 7 vps. he recieved 98 points for > that win. > thats more points than i got for going into the finals at a ninty player > tourney while i scored 1 more vp in one less game (these are just examples). > > all i am saying is that size should matter, and the size coefficient is > clearly not efficent enough. ... for some.

Daneel

> This is all true, sure. Before the current system went into place, I > argued for a scoring method that ran a little more linearly (or on a > curve, really) from 1st place down to 50th (or whatever last happens to > actually be), but what we ended up with is a system that weights making > the finals much more highly than anything else. Everyone here has read > the scoring system page, right? :-) Well, I see one controversy in that, and that is the fact that the finals always host 5 people, while the number of participants may be quite different from tournament to tournament. A 10-player tournament is going to have 5-player finals just like a 90-player tournament will. This actually means that in order to maintain a good rating, you have to go to small tournaments and win them; 1/8 is probably as hard as 6/88, and the latter is rated far lower. The VP and GW counting component is an excellent idea, by the way. For the other part of the rating though I could probably imagine something that has more to do with tournament ranking in general than giving bonuses for making the finals (which is further strengthening the differences between 5th and 6th place*). > I don't think this system is significantly flawed relative to many other > ways it could be done. I'm not trying to nitpick or anything, but you do realize that this is a fairly general statement... ;) > I do, as you say, think that it would make more > sense to slope the points earned down past the finalists more, but given > that tournaments have always been played with a "finals are the most > important game" system, it does make some sense for the finalists to > earn substantially more points than non-finalists. Yeah, whatever. I'm just giving my cents anyway, I mean, I wasn't there to contribute when the system was invented, so I don't really feel like having any "right" to complain now that its done. * I have an opinion about what being "good" means, and in my dictionary someone who scores 6th place three times in a row is better than someone who ends up in the finals once and scores zero VPs twice. I know excellent players with low ratings (myself not included) whom I fear far more as an adversary than most of the people who made the finals once or twice (out of luck, maybe, does not really matter). Bye, Daneel

Johannes Walch

> Yes, it is, by virtue of what it means to be a better finish. > > At the very least, it is open to interpretation and one of those > interpretations is that 3rd out of 40 is better than 5th out of 90. That is exactly the point. What IS a better finish? Very hard or impossible to decide. No matter how it is done, somebody will considered it flawed. The development of the current system was a very long process and took many ideas into account (one of them was not to piss off the playgroups which run many small sized tournaments just because they are far away from everybody else). I think the system is quite OK at the moment and it is way to early for another change, anyway. If you change the system every half a year or so people will quickly be annoyed... -- johannes walch

Johannes Walch

> This actually means that in order to maintain a good rating, you have to go to > small tournaments and win them; 1/8 is probably as hard as 6/88, and the > latter is rated far lower. Not true. As only the best 8 tournaments are taken into account you can go to whatever tourneys you like and still have a good rating. The No1 player has gathered all of his relevant points in 40+ player tourneys. -- johannes walch

Jeff Kuta

"LSJ" <vte...@white-wolf.com> wrote in message news:<2qteraF...@uni-berlin.de>... > "Stefan Ferenci" <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message > news:41497430$0$10840$3b21...@usenet.univie.ac.at... > > although i had a much better showing at a larger tournament i recieve 19 > > points less. > > A 3rd place finish is a better showing than a 5th place finish. How much better is a 5th place finish than a 6th place finish if andomization determines who plays in the finals? Seems to me if there are any ties that result in randomization being the final tiebreak, that the unlucky 6th place folks should at least merit points equal to a 5th place finish, even if they don't have the opportunity to improve. Otherwise the tie-break winners get a super bonus over the losers, both guaranteed Finalist Bonus RtPs and an opportunity to improve, perhaps win, which is HUGE for RtPs. Jeff

Daneel

On Thu, 16 Sep 2004 19:43:46 +0200, Johannes Walch <johanne...@vekn.de> wrote: > I think the system is quite OK at the moment and it is way to early > for another change, anyway. If you change the system every half a year > or so people will quickly be annoyed... > > -- > johannes walch I wouldn't like it changed too often either. Maybe we could let it run for a while, examine its shortcomings and then improve the system when a significantly better one is invented. Bye, Daneel

Frederick Scott

"Daneel" <dan...@eposta.hu> wrote in message news:opsefk5r...@news.chello.hu... > > I don't think this system is significantly flawed relative to > > many other ways it could be done. > > I'm not trying to nitpick or anything, but you do realize that > this is a fairly general statement... ;) I think what Josh is trying to say, here, is that it's kind of silly to quibble about the "fairness" of a system like this. A system based on granting players points for showing up at tournaments (whether it's JUST for showing up or whether you actually have to do something once you're there is ultimately irrelevant) is not fair in terms of rating players skills and never will be. Steve Wieck termed this a "rewards" system because it rewards players for participating and ultimately chose it because it's more popular (and quite possibly more understandable) than ELO or any other scheme purported to actually rate players' skill mathematically from demonstrated results. You show up at a tournament, you get some points for doing well. Your little number that you can go look up on the web reflects a set of these rewards point showings earned in the last 18 months. Enjoy. Have fun with it, for what it's worth. But let's not pretend one scheme or another makes more sense. Fred

Jeff Kuta

Stefan Ferenci <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message news:<41497430$0$10840$3b21...@usenet.univie.ac.at>... > i played in the ne qualifier this year and finished 3rd with 1 gw and > 5.5 vps. 40 player participated in the qualifier i recieved 98 points. > at gencon i played the final qualifier with 90 players participating. > i finished 5 with 2 gw and 8 vps but only recieved 79 points. I think it would be instructional if the results to several large qualifiers, with RtPs awarded, were posted on the WW Events site. Heck, you could even do *all* the qualifiers and championships. For the three continental champs and quals, that amounts to possibly 25-30 events all told? This would give people some sense of how much an event is worth. As it is now, it's completely arbitrary based on turnout. At least in That Other Game, you know how much an event is worth even before you show up. Some examples would demystify the ratings system somewhat. Jef

Stefan Ferenci

Johannes Walch wrote: >I think the system is quite OK at the moment and it is way to early > for another change, anyway. If you change the system every half a year or so > people will quickly be annoyed... > > -- > johannes walch > > of course you think the new system is ok, because you codeveloped it. and for the system not beeing there long enough, since we it was applied retroactivley, we have the data for 18 months and they show some flaws. nobody is asking for major changes, just some modifications like at larger tourneys (e.g.) extra points for all top ten players, at championship maybe eventhe top 15 to 20 players (lessening the difference between 5th and 6th place.) adjusting the size coefficent to make larger tourneys more valuable giving larger extra bonus to championships and qualifiers due to the more competitive environment. stefan

Daneel

On Fri, 17 Sep 2004 12:00:30 +0200, Stefan Ferenci <a950n...@unet.s.p.a.m.univie.ac.at> wrote: > Johannes Walch wrote: >> I think the system is quite OK at the moment and it is way to early >> for another change, anyway. If you change the system every half a year >> or so >> people will quickly be annoyed... >> -- >> johannes walch > > of course you think the new system is ok, because you codeveloped it. Talk about some fine print for ya... ;) > and for the system not beeing there long enough, since we it was applied > retroactivley, we have the data for 18 months and they show some flaws. > nobody is asking for major changes, just some modifications like > at larger tourneys (e.g.) extra points for all top ten players, at > championship maybe eventhe top 15 to 20 players (lessening the > difference between 5th and 6th place.) > adjusting the size coefficent to make larger tourneys more valuable > giving larger extra bonus to championships and qualifiers due to the > more competitive environment. > > stefan All in all the stuff does not really bug me. I am mildly aware of some inconsistencies, but since I couldn't give any less for the ratings in general, I consider it more of a theoretical issue. Ratings only count when you can objectively measure performance (and not just arbitrarily rank it). VTES is a multiplayer game and I don't really think that it is possible (or worthwhile) to measure skill in it. -- Bye, Daneel

Decebalus

> Stefan Ferenci <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message news:<41497430$0$10840$3b21...@usenet.univie.ac.at>... > > i played in the ne qualifier this year and finished 3rd with 1 gw and > > 5.5 vps. 40 player participated in the qualifier i recieved 98 points. > > at gencon i played the final qualifier with 90 players participating. > > i finished 5 with 2 gw and 8 vps but only recieved 79 points. I agree that bigger tournaments should be given more weight. But thats only an opinion. But your problem (20 pts.) seems to be irrelevant in a system where the best players have more than 1000 pts. There is really no evidence to talk about a "flawed system". Frank

Joshua Duffin

"Jeff Kuta" <jeff...@hotmail.com> wrote in message news:621dd332.0409...@posting.google.com... [ quoted text not captured ] I'm guessing that WW doesn't have the resources to do this on any kind of instantaneous time scale, but in the meantime, here's one sample event at least. These results are from the Washington DC qualifier tournament of June 2004. It had 33 players. prelim prelim final rating rank GWs VPs VPs points -------------------------------------- 1 2 7.5 1.5 155 2 2 8.5 0.5 134 3 2 7 0.5 81 4 2 6 0.5 67 5 2 8 0 63 6 1 7 41 7 1 5 33 8 1 5 33 9 1 4.5 31 10 1 4 29 11 1 4 29 12 1 3.5 27 13 1 3 25 14 0 4 21 15 0 2 13 16 0 2 13 17 0 2 13 18 0 2 13 19 0 1.5 11 20 0 1 9 21 0 1 9 22 0 1 9 23 0 1 9 24 0 1 9 25 0 1 9 26 0 0.5 7 27 0 0 5 28 0 0 5 29 0 0 5 30 0 0 5 31 0 0 5 32 0 0 5 33 0 0 5 Josh areas of competence: excel-to-text conversion

Reyda

"Daneel" <dan...@eposta.hu> a écrit dans le message de news:opsefcx3...@news.chello.hu... > On Thu, 16 Sep 2004 08:24:09 -0400, LSJ <vte...@white-wolf.com> wrote: > That should probably mean that he should gain more Rank Points for the > GenCon performance. "Besting" 85 players seems harder than "besting" 37. > Accumulating 2/8 seems harder than accumulating 1/5.5. (...) > Most representative of performance is perhaps the actual GW/VP scored, > regardless of any position. Or the number of players who were "bested". In > my opinion, of course. ;) > > Bye, > > Daneel Very good point Daneel ! but who can come with a good formula refelcting this ?

Kamel SENNI

Stefan Ferenci <a950n...@unet.s.p.a.m.univie.ac.at> wrote in message news:<414ab657$0$8024$3b21...@usenet.univie.ac.at>... [ quoted text not captured ] I love that one. The point is simply that a ratings system is made from "judgment' call'", arbitrary decisions made on a perception of the situation. For example : I play in an environment with a lot of tournament with 40+ players. I play four or five time in a year in tournament with 70+ people. I could prefer to raise the bonus for that type of tournaments. Now, for the people who play in a lesser size of player base, that sytem will sucks. Clearly, you're in the first case (and me too, stefen, and I've work wth Stephane Lavrut on this system, at least by spending my time to listen his thoughs all the time!!!;-)))(hello, Stephane !;-)). But think about the majority of the VTES community. This system would create a sort of star system based on the players who plays in the biggest player base. It would be as biaised as that you said, with the problem of let players in "small" town out of the competition. I'm now fourth wolrdwide or something in that system. I don't agree to raise my rank by let players from "small" town (in fact less that 20 players) out of the competition... Fo the personnal things, be more subtle, please: Johannes and I played in the type of "big" environnement I've describe. It would be better for US to have a system like you said. We would'nt. So a (little, ok..) attack on his envy to let the system live his life before changing it, for your estimated reason of be proud of that one (johannes have never be an official part of the development as me, except for the discussions between stephen, and us; and Steve Wieck doesn't said his name, in my memory, but stephane Lavrut's only...), is quite...ahem... well, i'ts not true...))). Hey, by the way, congratulations for your performance ! Kamel Senni, Prince of Paris.

Stefan Ferenci

just to clarify. i did not want to attack johannes in any way. Nor stephane, andreas or anybody else, who did their share in developing this rating system. (i do appreciate what they did (and still do)for vtes)i do prefer the new rating system and calling it flawed is maybe exaggerated, but i truly believe it needs some finetuning. if anyone feelt insulted by my posts, i do appologize. Stefan (btw kamel i really do appreciate your calm and very accurate style of posting)

Frederick Scott

"Reyda" <true_...@hotmail.com> wrote in message news:414af837$0$4034$79c1...@nan-newsreader-07.noos.net... > > "Daneel" <dan...@eposta.hu> a écrit dans le message de > news:opsefcx3...@news.chello.hu... > > Most representative of performance is perhaps the actual GW/VP > > scored, regardless of any position. Or the number of players > > who were "bested". In my opinion, of course. ;) > > Very good point Daneel ! > but who can come with a good formula refelcting this ? No one - because the premise is wrong. Actual GW/VP scored is NOT necessarily reflective of anything nor the number of players defeated. It all depends on who you scored your GWs and VPs against. And to take this into account, you need to start with a scheme that attempts to take it into account. Just counting game wins or victory points avails nothing. Fred

Reyda

"Frederick Scott" <dont...@mymail.com> a écrit dans le message de news:OWG2d.97022$yh.49119@fed1read05... [ quoted text not captured ] so what do you suggest ? unlike chess, vampire is not a duel type game and has a lot to do with luck. You can beat the #1 Vtes player if you sit at his table because of his poor drawing and using cross table interaction. You cannot beat the #1 chess player by luck or manipulating someone else. Does it mean the whole idea of rating vtes players should be flushed away ?

Frederick Scott

"Reyda" <true_...@hotmail.com> wrote in message news:414b8fbb$0$8722$79c1...@nan-newsreader-04.noos.net... > > "Frederick Scott" <dont...@mymail.com> a écrit dans le message de > news:OWG2d.97022$yh.49119@fed1read05... > > "Reyda" <true_...@hotmail.com> wrote in message > > news:414af837$0$4034$79c1...@nan-newsreader-07.noos.net... > > > > > > "Daneel" <dan...@eposta.hu> a écrit dans le message de > > > news:opsefcx3...@news.chello.hu... > > > > Most representative of performance is perhaps the actual > > > > GW/VP scored, regardless of any position. Or the number of > > > > players who were "bested". In my opinion, of course. ;) > > > > > > Very good point Daneel ! > > > but who can come with a good formula refelcting this ? > > > > No one - because the premise is wrong. Actual GW/VP scored is > > NOT necessarily reflective of anything nor the number of players > > defeated. It all depends on who you scored your GWs and VPs > > against. And to take this into account, you need to start with > > a scheme that attempts to take it into account. Just counting > > game wins or victory points avails nothing. > > so what do you suggest ? At this point? Just not worrying about it. I don't aim to dredge up arguments about ELO at this point, if that's what you're concerned about. I'm just pointing out that you have to do _something_ like ELO in order to have any hope of accurately rating player skill. > unlike chess, vampire is not a duel type game and has a lot to do > with luck. You can beat the #1 Vtes player if you sit at his table > because of his poor drawing and using cross table interaction. > You cannot beat the #1 chess player by luck or manipulating > someone else. Sure. I agree, these make it more challenging to construct a mathematical way of rating players in V:tES. Perhaps they make it impossible, I can't say. The one time we tried it, we chose coefficients that were brainlessly broken and no one paid attention or bothered to notice this for two years. (In retrospect, that was really dumb of us all, IMHO. I knew _something_ was wrong - but I just *assumed* whoever had proposed the system was bothering to watch how it worked. Guess that was naive...) It may well be that if we chose good ones, we'd get reasonable numbers out of the thing. But I agree, it doesn't seem likely we'd ever get anything as dependable as chess ratings. > Does it mean the whole idea of rating vtes players should be > flushed away ? I'm not sure what this question is getting at. If you mean, can we ever bang *THIS* system into one that will accurately rate player? Heavens, no! It doesn't even begin to do the right things. As I said in the other post, it's a *reward*-based system that gives you points (or the opportunity for points) as a reward for showing up. The basis on which it grants points to players is non-existant. Basically, we'll give some quantity of points to players that have done things that _sound_ good. Who knows if it's the right number? But why worry about it? It's a system for assigning bragging rights, nothing more. People probably find it fun to go look up their numbers after they've been granted some points. If that's some fun (and I won't deny looking up mine, either), then so be it. Just, let's not get too passionate about arguing what the "right" number of points for each acheivement turns out to be. Fred