The point system is busted.

[Shen]

Bullados, in the sentence immediately preceding that comment I already addressed that.

If you play Spiritomb, there's no way Sableye will have access to Poketurns. Your bench is virtually off limits to them outside of Seeker, and they will still need a Crobat and a Special Dark to KO your Spiritomb.

I never denied Spiritomb would be KO'd. I said your bench is off limits, meaning your opponent will not have access to Poketurns. If you open with Spiritomb and another basic, you should be okay. I realize there is no way to guarantee that you will open with 2 basics, but this is still more than what most decks can do. If you can survive the first turn from Sableye, you should be alright. As a preemptive counterargument, it is unlikely that they can Seeker your bench as well because they already need to play a Collector.

On the topic of Regice, it takes up one more card they need to get with Collector. They also need materials to discard for Regimove. Is it really that realistic to think they can knock out two Pokemon if they have to get Regice and discard 2 cards at the same time? Cards to get with collector include: Sableye (if non-Sableye start), Unown Q (if non-Sablye, non-Crobat start), Unown DARK, Uxie (to collect resources KO your benched Spiritomb once they move it), Crobat, and Regice itself. Facing a Spiritomb isn't too favorable for Sableye.

 

[InnoceNt]

Bullados, in the sentence immediately preceding that comment I already addressed that.



I never denied Spiritomb would be KO'd. I said your bench is off limits, meaning your opponent will not have access to Poketurns. If you open with Spiritomb and another basic, you should be okay. I realize there is no way to guarantee that you will open with 2 basics, but this is still more than what most decks can do. If you can survive the first turn from Sableye, you should be alright. As a preemptive counterargument, it is unlikely that they can Seeker your bench as well because they already need to play a Collector.

On the topic of Regice, it takes up one more card they need to get with Collector. They also need materials to discard for Regimove. Is it really that realistic to think they can knock out two Pokemon if they have to get Regice and discard 2 cards at the same time? Cards to get with collector include: Sableye (if non-Sableye start), Unown Q (if non-Sablye, non-Crobat start), Unown DARK, Uxie (to collect resources KO your benched Spiritomb once they move it), Crobat, and Regice itself. Facing a Spiritomb isn't too favorable for Sableye.

The thing is, it really IS. Sableye can take about 3 priizes off VileGar while they're setting up. It's only a couple late game techs away for winning the matchup easily.

 

[Shen]

Like I said, Sablock will be a beast, but I think Spiritomb lets you actually play a game against... albeit a losing one. If you don't run Spiritomb or Sableye, it's unlikely you'll have a game at all.

 

[InnoceNt]

boy, regs double byes sure do look REALLY good now...

 

[Swordfish1989]

In another TCG I played, the system was a bit different but I think it really worked. Here's a short summary:

The equivalent to Worlds was held twice a year, and to attend, you needed to have accumulated a total of 10 points (cumulative, your entire life).

the equivalent to what I guess would be Cities would have a Top 8 where 1st place received 10 points, 2nd place 5 points, 3-4 got I think 2 points, and 5-8 got 1 point.

the equivalent to Regionals would have a different system, where if I remember correctly, I think at least the top 4 got 10 points, and it cascaded down to I don't remember what number.

Also, they had a cutoff where I think the top 100 players in the world (based solely on ratings) got an automatic invite to worlds as well .

Under that system, I got 2 invites; one for being in the top 100 from my ratings, and then again from gathering 10 points over the course of I think a year and a half.

There was a max however - you couldn't accumulate more than 40 points at any given time.

I really like that this system keeps track of your performance from the beginning and it doesn't get reset. They keep it from getting stale at the top by having the point system, so anyone can gather points. What's also good about keeping your points forever, is that, each year Worlds was held at a different place, so it allowed me to keep my points while choosing where I could attend worlds. Just because I'm invited doesn't always mean I can go.

 

[MattPL]

You can only lose 32 points max per game in a K-32 event. And this thread seems like sour grapes to me. There are always tons of players who feel that they deserve a greater rating than they currently have.

The proposed solution destroys one of the most beautiful aspects of an ELO ranking system: the points in the system are always divisible by 1600. However many points you win in a game, someone else loses. This is how the ranking system self-regulates. Injecting extra points into the system throws it off balance.

There is no surefire way of determining player skill because Pokémon is inherently luck-based. The 40 top ranked players should not be the actual 40 most skilled players; you reach a point where differences in skill are subjective and very small. There has to be a luck-based way to make worlds.

Uxie donk is very easy to counter. Simply play Dialgachomp and you won't have to worry about it. Of course, doing so will put you at a disadvantage with regard to certain other decks in the format. That's how the metagame works. If you play Gyarados, for example, you deserve to be donked at least once... those 30 HP do mean something you know.

That's exactly what I was thinking, especially what is bolded. If you're ranked higher, then that obviously means you got somewhat lucky. You can't get a high ranking based on skill alone.

 

[SteveP]

@MattPL and Box of Fail: Absolutely correct. You win some by luck, and you lose some by luck. HOWEVER, I think that's what some of us are trying to suggest remedies for.

Is there a way to scale games that are won by luck differently than games that are won by skill? Both the winner AND the loser of those games would gain and lose points equally, keeping ELO a zero-sum system.

@Swordfish: The point system you explained seems very similiar to what Decipher used back when my kid was still playing Lord of the Rings. My kid hasn't played LOTR in over 3 years, but he's still one of the top-ranked players. IMO, at some point, you need to apply an aging factor to points. I too don't like the yearly reset in Pokemon, but it's better than no resetting or no aging.

 

[NoPoke]

yes there is a way to accommodate luck: see the picture for how the stake is modified when a Bayesian correction is applied to accomodate games that are a mix of luck and skill.

http://pokegym.net/gallery/showimage.php?i=51407

 

[SteveP]

yes there is a way to accommodate luck: see the picture for how the stake is modified when a Bayesian correction is applied to accomodate games that are a mix of luck and skill.

http://pokegym.net/gallery/showimage.php?i=51407

Yeah, but this doesn't actually see into the game itself. It just makes an assumption that a certain percentage of games are won by luck rather than skill. For example, a lesser ranked player might actually beat a higher ranked player by skill, but this makes the assumption that luck played a bigger role than it might actually have played.

If there was some way to weight games individually rather than use an extrapolation curve, that would be much more accurate. The Bayesian correction is a nice macro-level improvement. I was looking more at a micro-level improvement

 

[NoPoke]

the whole of elo is a long range estimation. It isn't possible to establish relative skill even in a game of chess based upon a single match. I don't believe that there is a micro level improvement that doesn't also come with a whole host of problems. Hence focusing upon the problems with an estimation system like elo where not all games are determined by skill.

 

[Porii Sames]

I don't mind luck in the system.

I guess it's because in general I'm a very lucky person...

 

[SteveP]

the whole of elo is a long range estimation. It isn't possible to establish relative skill even in a game of chess based upon a single match. I don't believe that there is a micro level improvement that doesn't also come with a whole host of problems. Hence focusing upon the problems with an estimation system like elo where not all games are determined by skill.

In addition to a K-factor for games, there could easily be an L-factor added to the equation. The L-factor would scaled the points, and be based on some objective game-stats like prize difference, less than X-turns (ie. 1TKO, 2TKO), etc.

I assume the curve you linked to takes into account that lesser-ranked players are more likely to win by luck than higher-ranked players, right? The problem is, that's not a viable factor because ELO is mostly a regional rating.

Anyway, it's definately a better measure for ratings that a straight curve. :thumb:

 

[Todd Ingram]

What I really hate about donks is that there are specific decks to counter it. I know this sounds wierd but think about it. It forces everyone to either use 'Chomp and other counters instead of a deck that has a chance at actually winning. Therefore half the players use donk decks and half use donk counters.

The only real way to solve the problem is to ban certain cards from current formats, but everyone knows that would be incredibly stupid and destroy a part of the game. Hopefully this will become less of a problem once DP and Plat get rotated out, but until then we'll have to trudge it out.

 

[LoTad]

The ratings system isn't perfect, but it functions. I feel like everyone on here is overreacting. If you tweak your deck's build, it should be fairly easy to guard against donk decks. Someone gave gyarados as an example of a deck that is easily donked because many of the basics have very low hp's. Idk about you guys, but my gdos list runs 19 basics. Only 5 of them have under 50 hp (karps and a q). So if the average hp of a pokemon is around 70, and I'm going to draw around 2.5 + basics in my first 8 cards, I should have about 210 hp on the field. That plus the 50 percent chance of going second and the chance of getting a mesprit in hand first turn makes the probability of donking fairly low. Donk decks suck but they aren't a significant problem for the tcg.

I'm surprised no one brought up the idea of an actual rule change instead of banning cards or changing the ratings and rankings system. Someone mentioned this a while ago on the gym, but what if the rules were changed so that if you don't have any pokemon in play, you don't immediately lose (or loose for like 20 people on this forum). What if you still had a turn to produce another active pokemon before the game was over. Granted, I haven't given this concept a ton of thought, but it seems like something like that would solve the entire donk problem for good.

 

[sToRm07]

I'm surprised no one brought up the idea of an actual rule change instead of banning cards or changing the ratings and rankings system. Someone mentioned this a while ago on the gym, but what if the rules were changed so that if you don't have any pokemon in play, you don't immediately lose (or loose for like 20 people on this forum). What if you still had a turn to produce another active pokemon before the game was over. Granted, I haven't given this concept a ton of thought, but it seems like something like that would solve the entire donk problem for good.

My thoughts exactly! Giving your opponent a turn to recover, instead of destroying them before they can even do anything just makes sense.I know all card games have an element of chance, but it is the mark of a good card game to have this element harnessed and not exploited like I feel happens in the PTCG all too often with T1 donks.

MY only qualm with this post - running a ton of basic pokemon is easy - if you are sp - but if you are not, making your deck more donk proof is a costly endeavor and could hurt your many other match-ups.

Think of it like this, a Luxchomp teched out for the mirror game excels against the mirror game, but is arguably not as potent against other giants in the format.

Sadly, SP is already basically much more immune to being donked than most other decks, no surprise there, but that is for another topic... isn't it?

 

[Sue Quan]

The point system is absolutely fine. if u get donked, thats how the game s and u wont loose more than 32 points right now. 32 sounds bad but at least ur able 2 recover from it from other games

 

[jeffrey123]

That's not the point I am trying to make. The fact is when you're at nationals and this is your last event, getting donked 2/10 games can finish your season. I know that for people who don't care about their rating and love to play just to ruin others *ewh, I just drooled* then it's really lame and that's what busts the point system.

If you think the point system is fine, you either

1) Don't play enough to care
2) Are one of the people I mentioned

 

[bullados]

It's one thing to jump up and down and say that the point system is broken.

It's quite another to suggest a potential improvement that is actually better than the system you're trying to replace.

Ian has produced a system that he's been looking at for a couple of years, and it looks like a pretty reasonable solution.

What is your solution to the problem that you see?

 

[Box of Fail]

That's not the point I am trying to make. The fact is when you're at nationals and this is your last event, getting donked 2/10 games can finish your season. I know that for people who don't care about their rating and love to play just to ruin others *ewh, I just drooled* then it's really lame and that's what busts the point system.

If you think the point system is fine, you either

1) Don't play enough to care
2) Are one of the people I mentioned

Uhm, well for any good player, points actually matter.

Deserves to get donked? Your point is now irrelevant.

I don't think you have enough credibility for me to listen to you. No offense

The attitude you're showing is one of a disgruntled player who is obviously one of the best players ever, but he might not make it to worlds because the format is so rigged against skilled players like him. Maybe if you had toned down the condescension a little we wouldn't have assumed that this thread is nothing more than a sob story about how you're scared you won't make it to worlds.

You seem like an OK player and a decent guy, but you don't ooze so much skill that you automatically deserve a trip to worlds. Nobody does. Being fairly good at this game doesn't entitle you to an invite. Plus, if I were making a list of people who should automatically earn an invite, you wouldn't be on it. I don't think you're a bad player, but you certainly aren't the crème de la crème you seem to think you are.

 

[Darkwalker]

The biggest issue I have with the current ratings system is that the best players are better off NOT EVEN PLAYING in the biggest events of the year (Regionals and Nationals) if at all possible. As an example, in my area Chris Fulop has to play the system to gather up enough rating points to garner an invite to worlds each year. What this entails is dominating cities and selectively dropping out of top cut when the other decks that make top cut do not look favorable to the deck he chose to play. His focus is not so much to win an event as it is to earn a certain amount of points at an event. I'm sure he is not alone around the country in being stuck doing this to preserve rating points. Herm Edwards has been famously quoted, "You play to win the game!". This is not the case in Pokemon. I don't know what can be done to fix this particular problem with the system we use, unfortunately.