Smogon Shoddy Server Statistics - January 2010

[EvilMario]

Originally Posted by Fat SHUCKLE MAN

I agree with you. In, RBY, there was perhaps about 15 common Pokemon in OU (Chansey, Tauros, Starmie, Venusaur, Rhydon, etc.). However, do bear in mind that there were less Pokemon in the game in RBY, so it was acceptable for there to be less Pokemon per tier.

In Gen IV, about 15 per tier wouldn't be enough, as it'd take forever to get down to the tier where Pokemon such as Sharpedo and Magneton are useful, let alone the tier where Luvdisc, Lickitung and Delibird can be used.

Likewize, in Gen V, we'll probably have to adjust the tier sizes, so they're not unrealistically small. There's no way we'll be having tiers of 40 if there's 650 Pokemon.

I'd say 600 is the more realistic number to shoot for. And of those 107 new additions, at least half, if not two thirds, will be unevolved forms. And out of that, about half might become OU/BL.

I wouldn't say tier size has to drastically change.

[Lockeness]

Just noticed that Latias has been getting a ton of usage on the suspect ladder. Hmmm time to prepare myself. Thanks again doug

[NightmareZ]

wow I'm glad to see jolt has started getting some extra usage as in the right hands that thing is deadly. And i have a question how exactly do you play starmie as i am having serious trouble using it.

Thanks, NighmtareZ

[Bologo]

Originally Posted by Fat NightmareZ

wow I'm glad to see jolt has started getting some extra usage as in the right hands that thing is deadly. And i have a question how exactly do you play starmie as i am having serious trouble using it.

Thanks, NighmtareZ

I personally use it as an Expert Belt lategame sweeper. Thunderbolt/Ice Beam/Hydro Pump/Psychic (or Grass Knot) gives such awesome super effective coverage, and Starmie easily has the Speed to be a great lategame sweeper. I think this is the best way to use it, because very few things get such great coverage combined with great Speed and good Special Attack. I don't like Life Orb because Starmie is vulnerable to entry hazards and Sandstorm/Hail, which means he won't be sweeping for very long, and Recover takes away coverage.

However, there are plenty of uses for it still. It's one of the better Rapid Spinners out there, because unlike Forretress (Payback is weak) or Tentacruel, Starmie gives Rotom-A quite a run for its money when it switches in to block Rapid Spin.

Also, as a dual screener, it works very well. Its combination of Speed and bulk along with a recovery move let it come in and set up screens more than once in the match, and STAB Surf or Hydro Pump can make the big Pursuiters think twice about switching in directly, especially since they won't be doing a huge amount once Reflect is up.

One thing you really have to make sure of when you're using Starmie, however, is that you get rid of ScarfTar. Unfortunately, ScarfTar is pretty much a guaranteed revenge-kill, or even a direct kill if it can lure a move that isn't a STAB Water move.

[Mr.E]

Another reason I don't like Life Orb on Starmie is that it's completely counterintuitive with Rapid Spin, for those of us using that move on our Starmie. WHOA BUDDY I DID 1 MORE DAMAGE... and did 10% to myself. :/

[wavedash]

I use Life Orb Rabid Spin Starmie. Depending on what your Starmie is doing, it's not a bad set.

[mtr]

Its a set with a very specific use: to cause serious trouble for stall teams. Starmie itself is useful against offensive teams due to its speed and coverage, but I prefer a defensive spinner because its also a decent Gyarados check.

[SHUCKLE MAN]

Originally Posted by Fat EvilMario

I'd say 600 is the more realistic number to shoot for. And of those 107 new additions, at least half, if not two thirds, will be unevolved forms. And out of that, about half might become OU/BL.

I wouldn't say tier size has to drastically change.

Yeah, we don't know for sure, but do bear in mind that we're also now treating NFEs as Pokemon in their own right now, as we've realized that using labels that Game Freak put on the Pokemon means nothing when it comes to competitive battling.

I think perhaps, as a guideline, we should be aiming to have about 10% of the Pokemon in OU. This sounds very arbitrary, but I also think it's common sense. If we still had tiers of 15 Pokemon, like in RBY, we would be on our 3rd usage tier, and we'd still have like 400 Pokemon which aren't competitively viable in any metagame.

[WarriorPrince]

Dragonite decreased. I was expecting him to crack the top 20.

[HellaJeff]

I am a little confused,
obviously with my post count you can tell I am pretty new, but I am here to learn.

What are these numbers telling me about the game? I understand that tiers are created based on what is being played on Shoddy but I have trouble understanding the impact.

I am most interested in playing OU or, starndard. I wanted to play this bracket because I thought it would have the metagame with the most freedom to take pokemon normally considered "out of tier" and build them in a team that could still rock.

But the way discussion on here goes it seems like the point of tiers is to create fine lines of usefulness instead of measure popularity and create fair boundaries based on that.

I hope that made sense and I would appreciate anyone who can help me learn the competitive game better. : )

[lordkira]

Originally Posted by Fat WarriorPrince

Dragonite decreased. I was expecting him to crack the top 20.

Same here.
Since I've seen uite a bit of (well some) Anti-Lead Nite and also a lot of offensive dragon dance variant with extremespeed. I guess maybe people are starting to see that it's not as great maybe?

[Deep Thought]

Originally Posted by Fat HellaJeff

I am a little confused,
obviously with my post count you can tell I am pretty new, but I am here to learn.

What are these numbers telling me about the game? I understand that tiers are created based on what is being played on Shoddy but I have trouble understanding the impact.

I am most interested in playing OU or, starndard. I wanted to play this bracket because I thought it would have the metagame with the most freedom to take pokemon normally considered "out of tier" and build them in a team that could still rock.

But the way discussion on here goes it seems like the point of tiers is to create fine lines of usefulness instead of measure popularity and create fair boundaries based on that.

I hope that made sense and I would appreciate anyone who can help me learn the competitive game better. : )

First of all, welcome to Smogon Jeff!

Now I will try to answer your questions because I can see where you're coming from and I don't want you to get the wrong impression of what's going on here. In addition to what I'm going to say, there are several excellent resources on Smogon's main site that probably explain things better than I could, so be sure to read those, starting with this one: http://www.smogon.com/dp/articles/intro_comp_pokemon

(1) The tiers in a nutshell: we start with standard. Pokémon that are too strong for this tier are banned to Ubers, where anything goes. It is technically a banlist and not a tier, but there is a recognized Uber metagame, ladder, etc. Pokémon that are used in at least 1 out of every 20 battles on the standard ladder are called "overused" or OU. The determination of the OU list is a little complicated, and I'm sure there's an article about it somewhere, but basically it boils down to a cutoff of ~3.42% usage on the ladder. Everything not in Ubers or OU is then eligible for UU or "underused" play - the same cutoff applies for the UU list and Pokémon voted too strong for this tier are put on the BL list. So the three main competitive metagames are OU, UU and Uber (there are only about 6 or 8 BLs so the BL metagame doesn't really exist).

(2) The server statistics - Again I will try to be brief, but there is lots and lots to discover here. Each month, DougJustDoug runs a program to take all the data pertaining to every single battle fought on the Smogon University server and churn out what you see on page 1 of this thread. You can see how many times each pokemon was used on a given ladder, what its common moves and items were, and pretty much everything else you might need to know about the status of the metagame. Obviously people will use what makes them win, so it's an objective view of what is "best" (although this is only true in a veeery general sense). Hence, what is most popular is popular for a reason, although there are exceptions *cough* Electivire at #34 *cough*. There's no reason not to use whatever you want to, but the stats are a great way of seeing trends in the metagame so you can try to make the best team possible, which will usually end up being 6 OUs in OU if you are really serious about reaching the top of the ladder.

To sum up: the tiers really do measure popularity (at least OU and UU do), but that's all they do. Use whatever you want - what often separates the good from the best is the ability to use unorthodox sets to "counter the metagame." Hopefully that clears things up a bit for you; if not, all I can say is lurk more and ask more questions!

[HellaJeff]

Originally Posted by Fat Deep Thought

First of all, welcome to Smogon Jeff!

Now I will try to answer your questions because I can see where you're coming from and I don't want you to get the wrong impression of what's going on here. In addition to what I'm going to say, there are several excellent resources on Smogon's main site that probably explain things better than I could, so be sure to read those, starting with this one: http://www.smogon.com/dp/articles/intro_comp_pokemon

(1) The tiers in a nutshell: we start with standard. Pokémon that are too strong for this tier are banned to Ubers, where anything goes. It is technically a banlist and not a tier, but there is a recognized Uber metagame, ladder, etc. Pokémon that are used in at least 1 out of every 20 battles on the standard ladder are called "overused" or OU. The determination of the OU list is a little complicated, and I'm sure there's an article about it somewhere, but basically it boils down to a cutoff of ~3.42% usage on the ladder. Everything not in Ubers or OU is then eligible for UU or "underused" play - the same cutoff applies for the UU list and Pokémon voted too strong for this tier are put on the BL list. So the three main competitive metagames are OU, UU and Uber (there are only about 6 or 8 BLs so the BL metagame doesn't really exist).

(2) The server statistics - Again I will try to be brief, but there is lots and lots to discover here. Each month, DougJustDoug runs a program to take all the data pertaining to every single battle fought on the Smogon University server and churn out what you see on page 1 of this thread. You can see how many times each pokemon was used on a given ladder, what its common moves and items were, and pretty much everything else you might need to know about the status of the metagame. Obviously people will use what makes them win, so it's an objective view of what is "best" (although this is only true in a veeery general sense). Hence, what is most popular is popular for a reason, although there are exceptions *cough* Electivire at #34 *cough*. There's no reason not to use whatever you want to, but the stats are a great way of seeing trends in the metagame so you can try to make the best team possible, which will usually end up being 6 OUs in OU if you are really serious about reaching the top of the ladder.

To sum up: the tiers really do measure popularity (at least OU and UU do), but that's all they do. Use whatever you want - what often separates the good from the best is the ability to use unorthodox sets to "counter the metagame." Hopefully that clears things up a bit for you; if not, all I can say is lurk more and ask more questions!

Thanks a lot deep!

I guess it was just the discussion going on that confused me on whether or not you could be successful with a team that was not the top 15 on standard, but after reading your explanation I can see how several pokemon could be on teams that are on top of the ladder but would not have influenced the stats. Unless, the use of these pokemon were noticed and their strategies copied by other players. Once again thanks for helping me understand!

[SHUCKLE MAN]

A lot of people do want weighted statistics, but I think the raw figures are better for calculating the tiers. A lot of people don't have high ratings (like me), so our battles wouldn't have as much of an effect on the stats at the end of the month, which sucks.

But the main problem with that would be the fact that the lists would be more useful for the good players than for the normal players if they're based off weighted stats. The way it is now, the lists are useful for nearly all the players.

[Quaily]

I think the desire for such a list is because of curiosity, not to determine teiring because we know the lists of 1300+ 1400+ 1500+ would look radically different. Ninjask would probably be nu

[ultimifier]

tiers should be dependant on what people use when they are trying to win.

if some fuckhead runs around with a team of charizard, electrivire, gyarados, porygon2, ninjask and weavile and has an average rating of 1000 and plays 500 games a month that is really screwing up the tiers.

point is the tiers SHOULD BE based off players with 1300+ CRE (or some other arbitrary amount) as it weeds out the people that don't really take the game seriously.

[Stylish Interval]

Originally Posted by Fat SHUCKLE MAN

A lot of people do want weighted statistics, but I think the raw figures are better for calculating the tiers. A lot of people don't have high ratings (like me), so our battles wouldn't have as much of an effect on the stats at the end of the month, which sucks.

But the main problem with that would be the fact that the lists would be more useful for the good players than for the normal players if they're based off weighted stats. The way it is now, the lists are useful for nearly all the players.

Actually, if the normal players could see what the good players are using, and how they got up that high, they could probably improve themselves by ruling out the poor shit and improving their teams. I mean, if Michael Jordan got to the NBA through practicing free throws, rather than three-pointers, shouldn't you practice free throws to get to the NBA? (not-so-great example, but you get my point.) If a high ranked users used Specs Latias, instead of LO Latias, shouldn't the players take note of how much better it can do and use it to?

Just my thoughts on the subject that appears to be at hand!

[capefeather]

I think that one good, objective way to satisfy what ultimifier suggested is to gather statistics for people who actually DO win. If someone's playing around and not trying to win, he/she will naturally not win as much as a competitive player would, even if he/she plays a lot.

[ultimifier]

but then when two crappy people play around someone crappy still has to win. look at all those "warstories" that are two inneficient battlers playing. Every team has at least 3 of the following pokemon: electrivire, weavile, breloom, porygon z, heracross. Granted some of these pokemon are not useless (weavile and breloom) but A LOT more bad players use them then good players.

[Rhys DeAnno]

I'd be interested to see some of the bigger differences between raw usage and 1400+ usage statistics. From the games I've played, it seems like Ttar has a choice scarf at least 50% of the time on higher rated players (as opposed to less than 25%).

[capefeather]

I don't think that that matters that much. Bad players will still have lower win rates than good players will. Even with the similar-rating matchmaking, bad players will be stymied repeatedly by good players. I think that this would be a better solution than a discrete, arbitrary rating cutoff that may throw out stats from good players that don't ladder very seriously.

[Stylish Interval]

Well, if good players aren't laddering seriously then they don't need to be considered on the ladder. I mean, so what if 10-20 people are using BandTar when 300-400 people are using ScarfTar? It's a minuscule number that will not appear, aside from "Other Items" or whatever it says. It really won't show anything other than more Pokemon being used, but I'd rather know what the people at 1300-1400+ are using, not the people >1200.

Should I consider those few people threats when they aren't at the point on the ladder where I am? (Not saying, of course, that I'm 1300-1400, just that if I was I wouldn't care about those lower people.) I really should only consider the best players that are also where I'm at so that I know what to prepare my team for.

[capefeather]

Well, I was just trying to propose a solution assuming we wanted "good player" stats. The current system of collecting everything is quite useful for pure laddering purposes nonetheless.

[Mr.E]

A weighted list would be useless since 90% of the highly-rated players are still scrubby anyway. Ladder rating doesn't mean a whole lot.

[Purple Wobbuffet]

| 32 | Deoxys-l | 590 | 2.42 |

A ton of OU pokes are more used than it in Ubers. Potential suspect?