np: OU Suspect Testing Round 3 - So Long and Thanks for all the Fish

Status
Not open for further replies.
Consistency in what? It's consistent enough to win more than everyone else, and you're already putting in some serious playtime if you're parked at the top with an impossibly high score. Take some time off. If you come back and start losing, it will show in your rating and you won't be as consistent as you were when you won all of your battles. Rating decay just pressures you into playing obsessively to keep a high score. It doesn't quantify consistency at all.

Are there any other games you know of that use a system of rating decay? This is the first I'm hearing of it.
 
No, we are rewarding consistency.

No you're not, you're assuming that it means that people will be laddering all through the testing period. What it means in reality is that people will spam the ladder the last few days of testing, because that's the only time which your ranking metters - when this round officially comes to a close and phil takes that screenshot of the leaderboard. It doesn't reward consistency at all, you're just specifying the time frame in which you need to achieve the ranking. That's fine if you're flexible with your time, but not everyone can guarantee that they can play on the ladder a lot during the last 2-3 days of testing.
 
You only need to have one battle every two days for your ranking to stay constant. I think anyone can manage a battle per two days for a couple of weeks.
 
Haven't played yet, but there's probably not going to be anything else than HO Sand teams, experimental failing Sun teams and the usual gimmicks (which should be the normal teams used imo)...

Well, good luck to everyone. I'm gonna try my best.
 
I don't like this rating decay. It's hard enough to get to the top in the first place, but having to battle continuously for up to a month to stay there? The risk of getting a stroke of bad luck and losing it all? If you reach the requirement, you've qualified, or do you have to reach the requirement at a certain time?

Parking accounts IS a problem, but I much preferred Philips wrath to a rating decay....
 
Who cares about rating decay either man up and stop complaining or don't play at all, will make all of our lives easier and its not like the rating decay is huge anyway.
 
Have a look at the voter identification thread, where there is a screenshot of the ladder at midnight: http://www.smogon.com/forums/showpost.php?p=3255544&postcount=1

That is telling me that there is a problem with parking accounts at the cut-off. 44 out of 47 people making it being within 20 points of the limit (one battle, once you get that high) and then a few way off shows that people are stopping once they get there. You can't say that you will rely on 'the wrath of Phil' to stop it because he clearly hasn't, and Rating Decay will a least stop people only having to be above it once over the course of a month, because they will instead have to do it at a given time, which means people who can't do it consistently will not manage it. The variable cut-off will also help, admittedly, but decay will be useful to stop people parking if they get there once on a fluke.
 
Didn't Shoddy also have a rating system which made one's rating decline with time too? As I recall, one's volatility increased the longer they didn't play, and one's rating decreased with the amount of volatility their rating had, but I'm not sure if I fully understand it. The bottom line is: Shoddy also had a form of rating decay, so this system actually isn't so different.

In fact, the annoying part is that the new rating system sets the minimum in Find Battle variation to 200, which creates a whole bunch of +8,-23 battles once your rating gets high enough. Before, you could set the variation to something like 50, and get normal rating changes after battling.
 
I believe rating decay is essential. I reached 1350+ last time around, however i couldn't find ANYONE to battle with a 50/100 rating difference because they parked their alts in the 1400s or even the 1300s and never played again. I had to put the difference to 300 max to find a battle a day i was bored, lost to hax, and this brought my rating to 1317 immediately. If rating decay isn't enforced, members won't ladder after they've reached a sufficient record (or use a different alt), and this doesn't allow newer players to climb up the ladder. It's not hard to find 5 mins every two days to ladder anyways.
 
In fact, the annoying part is that the new rating system sets the minimum in Find Battle variation to 200, which creates a whole bunch of +8,-23 battles once your rating gets high enough. Before, you could set the variation to something like 50, and get normal rating changes after battling.

That is the most annoying part of it all, i used to set the variation to 100 just to stop all those haxy -23 battles... but that seems unavoidable now :/.

Oh well, no use complaining i guess. Time to pray to the hax gods.
 
Now, I want to make clear that I'm supporting the concept itself of rewarding playing a lot, not the rating decay itself. I actually don't think that the rating decay solves anything. I liked how Shoddy Battle's rating system increased deviation every day, addressing the real problem of accurately determining rating. The CRE decay was just a byproduct of that and not this huge selling point that PO's rating system is making out of its rating decay.

I also want to make clear that I'm talking about rewarding frequent, consistent play, not punishing people for not playing. I just think that there's a problem when a rating system doesn't directly recognize at all that I invested x amount of time into the suspect test unless it's an obscene amount or I anti-metagamed really well. If you're good at blitzing the ladder, fine, get your check, you deserve it. But people who spend the time testing multiple teams, some perhaps even not that great (maybe leading to some "do not ban" votes), deserve that check, too.

I'm sorry if this is deemed "unnecessary" to the kind of discussion that's wanted in this thread, but I just had to clarify my position on this matter.
 
I don't see the point. Who cares if someone doesn't get to play someone else because they parked, wasn't the whole goal of this is that voters get accurate experiences with the metagame and can then form an educated opinion of it afterwards? If you reached your 1400, you played in the metagame, and your obviously competitively up to snuff. Maybe you can't contribute well to a discussion but you clearly played enough and we're proficient enough in the metagame to know what you were saying.

Rating decay doesn't necessarily mean you will try new things or that you will face new things. So really the knowledge of threats doesn't necessarily change whether it's 15+15 or 1500 req or 1400 req.

Doesn't affect me though, I've been playing like 1 hour a week on average lately. so 15+15 or 1500 I don't have time for the laddering.

We've been playing this meta for a bit now anyways considering the changes happened a while ago.
Where are the opinions on the current state of this meta?
 
@All: Actually the settings for rating decay are perfectly adjustable (I don't know what settings Smogon chose) but if you consider the req as the average level you need to reach, then battling once every two days isn't death and you should be able to have an account above the req before the deadline, you just need to put your rating at what your average level is at least once during a two days timespan before the deadline, you'd be extremely unlucky if you can't even achieve that.

Or otherwise maybe you're just not good enough to qualify as a suspect voter.
 
My complaint is not the system in place, simply the lack of a reset. Player x, who ended round 2 above 1500, literally only has to battle twice in round 3 in the last day of the test to remain above that level and all but guarantee voting rights. Player y, already above 1400 at the end of round 2, similarly has very little actual laddering to do to remain in a position likely to allow them to vote.
 
My complaint is not the system in place, simply the lack of a reset. Player x, who ended round 2 above 1500, literally only has to battle twice in round 3 in the last day of the test to remain above that level and all but guarantee voting rights. Player y, already above 1400 at the end of round 2, similarly has very little actual laddering to do to remain in a position likely to allow them to vote.

The ladders get reset when the Test Round begins.
 
I find it mildly amusing that the biggest controversy in suspect this round is the ladder and ratings system. Why don't we move the discussion back to were it should be, the discussion of which pokemon deserve to be nominated for suspect. The rating system can be discussed either in PR or in its own thread.
 
@ Texas: Agreed, how about we stop moaning and start playing?

Anyway, I've played quite a few games today.

I've been doing rather well with my Drizzle Offense team. It probobly needs a tweak or two, Rotom-W isnt pulling its weight in my veiw, compared to the other members.

Other team types I've seen:

Sand: Rather common, but not lost to a Sand Team as of yet in Round 3.

Rain Stall: Considering it's built strong against Rain teams, I lost, but thats only because I mispredicted once and Parasect got in on an NVE move.

Sun: I've faced three Sun teams, although, none with my Rain team. It decimated my No weather team. That said, my No Weather team hasn't been doing that good, and is still in alpha. Oddly enough, Venuaur wasn't the problem, Heatran and/or Ninetales were, because my team was Fire weak.

Boosting Offense/Bulky Offense: No problems here. Renkulus and Conkeldurr have been the most annoying, but no issue.

Hyper Offense: These are the the which keep throwing powerful blow after powerful blow at you. Draco Meteors from the Likes of Lati@s or Sazandora every three turns. This is where I've seen the problems.

What seems broken now?

This is based on my Rain team, which is doing well, not my No-Weather team, surprisingly, there isn't a lot which strikes me as broken, although there may be Wobufett as well, but I've only faced a single one.

Latios: I know, it's been a suspect twice already. But in Round 1 and Round 2, the Hyper-Offensive metagame and the sheer insanity of Rain and Sand being everywhere kept Latios held back a little, as DM was a free invitation for Kingdra/Kabutops/Dory/Landos to come in and wreck stuff afterwards. Now that SS is more or less gone, and Sand is actually used less, due to the smaller need to counter Rain, Latios is not held back at all.

Specs Draco Meteors take 50+% off from even Steels like Heatran and Jirachi. They're your best bet, besides something like Chansey/Blissey, at taking these blows. Basically anything that isnt a pink blob of fat is 2HKO'ed at least.

Then there's the Life Orb sets, which hit hard as anything... AND can change moves. Did I mention Recover?

It's particually nasty when Latios is paired with Sazandora or Latias. You simply cannot withstand that many Draco Meteors, unless you're a Chansey/Blissey Softboling every turn... and even then, a Crit will mess you up badly.

Basically, in short, Latios destroys Offensive teams, especially with another D-2 Bomber by it's side, such as Latias and Sazandora. Stall dosen't stand much of a chance either, with the possibly of Trick, or a Bulky CM set.
 
Hmm, clearly coyotte is confident in his rating system. Oh, well... What I think I'm going to do is summarize my experiences here regularly, maybe every day. I'm seriously going to prove myself in any way I can. Hell, I'll even give the rating system the benefit of a doubt by seriously not looking at my rating at all and seeing how the (hopeful) resulting removal of such a source of demoralization will affect my ability to qualify. I'm just regretting not being able and/or motivated to play in the past week that the banlist was available.
 
My complaint is not the system in place, simply the lack of a reset. Player x, who ended round 2 above 1500, literally only has to battle twice in round 3 in the last day of the test to remain above that level and all but guarantee voting rights. Player y, already above 1400 at the end of round 2, similarly has very little actual laddering to do to remain in a position likely to allow them to vote.
If it's the same settings as the default one, then they need 10 battles, not two, after a long period of inactivity. Due to them searching with the real rating (ie players in 1500 with +/- 200) they'd have to win 10 battles in a row against 1300+ players (and probably some 1400 and 1500s as you can't control who you face) so if they can pull off those 10 wins in a row once everyone's adapted to the metagame they definitely deserve the voting rights.
 
Status
Not open for further replies.
Back
Top