Playing Style and overcoming the Learning Curve to become "the best"

[Tangerine]

Every game has a learning curve, and of course, it is a part of the philosophy of Smogon to give a people an idea of what the learning curve is, and to aid them overcome the learning curve.

Obviously, I don't consider myself at the top of the learning curve yet - so I will be asking questions that I feel are relevant to the specific milestones that I pick out of the learning curve.

Thus, this thread shall be about identifying key points in the learning curve, and methods to overcome it. This is a discussion that would be most relevant to tutors, I would assume. This thread is essentially asking, "How do YOU play the game?" Meaning this topic will be about a lot of things (but it wont be about Garchomp, or Wobbuffet, etc) Of course, I'm but one person and I never had experience tutoring yet - so I'm sure many other people can point out on any key points and concepts that I have forgotten to cover.

I think we all pretty much agree - that Pokemon is a game of statistics. It is a game of managing these statistics and trying to win - or defeat all 6 of the opponent's Pokemon. We call it the game of risk management. We realize then that the goal is not to "win every game", but to "win as many games as possible".

Then the first step to learning Pokemon is clear - memorizing data, being familiar with the opposing Pokemon. Knowing how much damage your sweeper can do to X. Know what your opponent can do, familiarize yourself with the Pokemon in the metagame, familiarize yourself with potent combinations and strategies used by battlers.

The first step to learning Pokemon is then to "know the data", allowing yourself to use these statistics to manage your game. It is obvious that in a game of managing risks, information is clearly the most important aspect. While people can say that statistics doesn't matter at all, but everyone does this instinctively, whether they like to admit it or not.

Then, how do we manage these statistics? Where are the lines that should be drawn, and of course, I am asking the opinions of the players. Is there a "best" method to approach each of the questions?

1) Scouting.

The art of getting information about the Pokemon. This is simply about taking note of % damages, the items Pokemon are using, the moves it uses, and the combination of a Pokemon on their team.

How far can we take this concept? Would it be feasible for someone to assume that their opponent is a skilled team builder - and start making assumptions about the last few pokemon based on the moves used and the pokemon that shows up? Would this be a valid management for statistics?

2) Prediction and Risk management.

Being able to "read" your opponent's move and acting accordingly. When does one choose to "overpredict", and when does someone choose not to? I noticed that there are always the safest options when you can choose to overpredict (CB Hera up against Blissey can probably overpredict without serious repercussions) and there are times where you shouldn't even bother to overpredict at all (Deoxys E up against TTar - the incoming Superpower is obvious, but if you Ice Beam, predicing the switch to its' Gliscor, you lose).

Thus there is a risk management factor that needs to be dealt with. I heard stories about good battlers using HP Rock on their Pokemon in ADV because of the accuracy issue - does this mean that being conservative is the best approach to the game? When is "overpredicting" acceptable?

3) Surpassing the concept of Counter - forcing other players to deal with you, rather than you dealing with them.

Because the D/P metagame is one such that one cannot deal with all possible threats, it is important that one learns the ability to play around specific threats since there is no one single team that can deal with every threat.

However, I feel as if the concept of Counter has restricted a lot of people - that one "must" be able to counter a threat or else the team is next to useless. MoP's post illustrates this point well - how an offensive team is completely able to get around the big threats.

This is a playing style issue of course, and an opinion that differs from player to player. Then, the questions are, "What are the ways to deal with a Pokemon, without carrying a specific counter for the Pokemon?" How important is this concept?

4) Instincts

A lot of players tell me that they just go by their "gut feeling", or their instincts, and not about "statistics" (this reminds me of course, about the "Finish it off with False Swipe!" Anime episode. COULD THE ANIME BE RIGHT?!?!?!).

Someone told me of an experiment done by a statistics professor - who supposedly built the "perfect team" in terms of numbers, yet the team was terrible when applied.

The question is now, is Pokemon about managing risks instinctively, or is it capable for one to simply think everything through concretely without resorting to instincts? If the latter is possible, then what should be the template thought process, if that exists? Another way to answer this question is "Is there a Pokemon AI that can beat every other team"? (Colin has claimed that he has created one, but we never saw it in action so w/e)

5) New sets, Gimmick Sets.

I have defined Gimmicks as a "failed new set" - a new set that failed to be a trend. I notice that some people, some even tutors, have a love for little gimmicky sets that have potential but never take off since they are not very effective sets in the very end.

I also notice that people clamor for "novelty", attempting to become unique. This thread reminds me of a lot of things.

http://www.smogon.com/forums/showthread.php?t=39701

So, where is this line? How much importance should be placed in "gimmick sets", and attempting to utilize them? Should they take place in the learning curve at all, or should we simply encourage people to stop attempting to try?

How about new sets/Pokemon that "stay ahead of the Metagame"? being able to know what is current in the metagame, and making a set that takes advantage of the metagame until they become trends? If a player is new, when should they attempt to do this?

6) Strategies behind Team Building.

We notice that many teams don't have much of a strategy other than, "Counter things, whitter things down and wait for an opening and set up X to sweep". Many teams center around this strategy.

What are the role of Theme Teams? (Rain Dance, Sunny Day, Trick Room, Offensive), in contrast to this? Is there a certain strategy that always beats another strategy?

Can you define Synergy anything past "how well a combination of Pokemon work together to accomplish a task"?


7) "The Best"

If this game is about statistics, is there a "Best" way to play the game? If this game is just about statistics, then we could theoretically come up with the best way to accomplish this. While it's clear we don't have all the data we need, do you think that this is actually feasible? Or will it always simply be "rock paper scissors" on which playing style beats which?

Obviously I haven't covered everything - but I feel as if this has enough discussion worthy content... for now. I will update the OP as needed.

 

[darkie]

I do agree with you in that the first step in becoming a true 'pokemon master' is to know the data. Someone who just rushes into battle without really knowing what they're using and facing will usually lose. I know this from experience. Only once one has learnt the magic numbers, mechanics, and basic info about how the game works can he/she excel.

 

[Syberia]

I find pokemon to be largely a game of adaptation. I always start with a "team" of 6 things, with a rough idea of what I want it to do. But since nothing ever works in reality the same way it works on paper, I playtest the shit out of it. If something just isn't working, or I find myself constantly losing to a particular threat, I make changes accordingly. Sometimes it's changing the way I play, sometimes it's changing a move, sometimes it's changing a pokemon. In the end, if you lose simply because your opponent has pokemon X and you don't have anything to beat it, it's probably the best way to lose, because it probably tells you the most.

 

[kd24]

Man what is there really to say? You really covered everything and its hard really to respond but I will focus on 1 thing.

- getting to know the metagame.

I have to be honest here, when D/P came out I was simply overwhelmed. I felt like I was going to fail so bad in it after coming out from advance. I saw a lot of people move to D/P that I knew from netbattle such as taylor, gormenghast, ipl, ect. and a few people I had seen that were stars in netbattle (husk/vil/mystica). When they all left I took it upon myself to literally engulf myself in D/P play. I took hours at a time learning all the new items, abilities, moves, and pokemon inside and out. I felt it was like a responsibility for me to become one of those few "skilled" D/P players. I dont really know why but I felt like I had to. So after all this studying I downloaded shoddy battle (I didnt feel comfortable with it until I learned all about D/P). Now while most players who got shoddy played right away to test it out and see what was happening, I didnt. Instead I spent hours watching battles of all the vets and players I knew. I chatted with people and took it upon myself to learn this metagame inside and out, until I could control it myself. So I made a team after tons of researching and scouting and started playing. And it all payed off. I was phenomenal when I started simply because I had a feel for the metagame without even playing it. Now simply from playing everyday and writing down blogs of the trends I see I quickly adapt to the new stuff everyday working as hard as I can to be the best at it. Its pretty cool.

 

[Judgement]

"How far can we take this concept? Would it be feasible for someone to assume that their opponent is a skilled team builder - and start making assumptions about the last few pokemon based on the moves used and the pokemon that shows up? Would this be a valid management for statistics?"

I think that it would be feasible for one to think that their opponent is a skilled team builder because a skilled team builder will have Pokemon that work well together, fill roles as needed, and generally cover each other with resistances or stats. It definitely can work for guessing what other Pokemon an opponent may have; I have successfully predicted certain things in the past based on the moves that I am using, the moves that the opponent is using, and the Pokemon that I have seen because the logical choice for the opponent in a certain situation was to go to a Pokemon with an appropriate resistance or stat that could defeat my Pokemon. Then I could just make the appropriate switch at the same time and force another switch and keep it going until I am able to make a good move.

Let us say that you have a team that happens to have Gengar and Infernape. You have Gengar out on the field, and your opponent has Gyarados. The logical choice on your opponents part would be to not leave Gyarados in because Gengars tend to have Thunderbolt. You know that they will probably switch to something that resists or can sponge a Thunderbolt. Therefore, you could either a) use Thunderbolt anyway, b) use a different attacking move, such as Focus Punch or Shadow Ball, or c) use a status-inflicting move, such as Hypnosis. Just by having Gengar out at the same time as Gyarados, you can predict a special resist even if you do not know whether or not your opponent has one.

"Thus there is a risk management factor that needs to be dealt with. I heard stories about good battlers using HP Rock on their Pokemon in ADV because of the accuracy issue - does this mean that being conservative is the best approach to the game? When is "overpredicting" acceptable?"

Being conservative is not necessarily the best approach to the game because of the risks and rewards associated with predicting. I mean, it can be useful if you want to definitely hit almost every time (barring something like Sand Veil or something like that), but it may prevent you from taking the upper hand. A lot of the more powerful attacks have worse accuracy, and while it would be detrimental for you to miss, it can pretty much ensure a kill if it hits. I think that one should be conservative only if one's strategy or situation would absolutely need the improved accuracy.

From my experiences, overpredicting is fine when the negative outcome is less likely to occur or is less damaging.

"How important is this concept? [counters]"

It's both important and unimportant. As you said, MoP has shown how an offensive team can overcome threats without a specific counter. However, it can also be important by covering the one or two things that the rest of the team cannot. If your team can cover everything except, say, Tyranitar, then by all means have your 5 Pokemon and a Tyranitar counter that fits well with your team, because you want to win and you have to do what is necessary. But if you have enough resistances and what not to deal with a good number of things, then you don't have to worry about counters.

"The question is now, is Pokemon about managing risks instinctively, or is it capable for one to simply think everything through concretely without resorting to instincts? If the latter is possible, then what should be the template thought process, if that exists?"

It's both, because sometimes your mind knows more than you do and if you just follow your instincts, you might just make the one right move that you didn't think of. I think that it's impossible to go through a whole game without using your instincts, because sometimes you just know.


"So, where is this line? How much importance should be placed in "gimmick sets", and attempting to utilize them? Should they take place in the learning curve at all, or should we simply encourage people to stop attempting to try?"

The line is your team. If a "gimmick set" does what your team needs to do and does it effectively, then use it. If another Pokemon can do the same thing, but better, then use that better Pokemon. Only use gimmicks if it's exactly what you need. I think some importance can be placed in using them because it shows that one is trying to solve a problem differently. They shouldn't take place in the learning curve until much later, when one has a better feel for the game and how it floes. We shouldn't encourage people to stop attempting to try because they can work when used effectively.

When I was playing Advance (which I knew a lot better than I know DP now), I tried some "gimmicks" and they did work most of the time because I simply built around them. That's really all it takes and the results can be interesting.

"How about new sets/Pokemon that "stay ahead of the Metagame"? being able to know what is current in the metagame, and making a set that takes advantage of the metagame until they become trends? If a player is new, when should they attempt to do this?"

Staying ahead of the metagame is always good because it's taking advantage of the current trends and benefiting from them. If it becomes a trend, then just keep staying ahead because that will always work. Think about Advance, when the "Tyraniboah" set was first being used; it took advantage of the metagame until itself became a trend.

New players should do this when they understand the game pretty thoroughly and are able to identify the trends.


"What are the role of Theme Teams? (Rain Dance, Sunny Day, Trick Room, Offensive), in contrast to this? Is there a certain strategy that always beats another strategy?"

Their role is to use a defined strategy and execute it instead of just figuring out what an opponent has and countering them. They require just as much, or more prediction to use. Theme teams are pretty different than most other teams but they are still based on the same idea. However, they can be easier to defeat once an opponent figures out their holes and predicts correctly.

"Can you define Synergy anything past "how well a combination of Pokemon work together to accomplish a task"? "

It's really hard for me to define it in a way other than that, but it's worth a shot.

The definition of synergy is "the phenomenon in which two or more discrete influences or agents acting together create an effect greater than that predicted by knowing only the separate effects of the individual agents," so in the case of Pokemon, it could be taken to mean that not only do they work together to accomplish a task effectively, but they also do it in a way that makes each Pokemon more effective in the team than as a standalone piece in the game. One batter doesn't bring victory to a baseball team; everyone has to contribute, both offensively and defensively. It's the same in Pokemon. Each Pokemon has to be able to contribute in some way that they can help each other out and win.

"If this game is about statistics, is there a "Best" way to play the game? If this game is just about statistics, then we could theoretically come up with the best way to accomplish this. While it's clear we don't have all the data we need, do you think that this is actually feasible? Or will it always simply be "rock paper scissors" on which playing style beats which?"

There really isn't one good way; there are many styles of play and it depends on the player. Each player uses statistics differently, so I don't think that there is one "best" way to play. It's the same reason for so many different team combinations working. The metagame is balanced enough that it is impossible to really defeat everything. Even the best teams and players lose at some point. I don't think that its favorable because there will always be a weakness. There's also a certain amount of luck involved in Pokemon. The inclusion of luck means that statistics won't always work. Different ranges of damage, critical hits, misses, and simple human error will prevent statistics from working every single time. A statistically perfect team won't work every time because the player isn't perfect and they won't always be lucky.

In terms of my experience playing the game, I try to find different styles of play to exploit weaknesses. I've used many kinds of teams and I've been playing for a while. I take risks even if they don't always work out because I try to make educated guesses and I calculate moves. However, if the situation calls upon it, I go with my gut feeling because it tends to be right. Regarding the team building, counters, and what not... In Advance, I had a very good team that could beat just about anything, but it still had weaknesses. However, it is possible to work around those shortcomings, but if the opponent is just simply smarter or luckier, then it's hard to come out on top. I try to adapt and change my team as things change. I also watch a lot of battles to see what people use and what strategies are effective.

I believe that you've addressed a good amount of what one usually finds when learning this game. It's about using all of those things in concert to be an effective player. Some find it harder than others, and when I tutor, I try to work more on skills that people have the least experience in. It's really just a matter of knowing what is happening.

I learned how to play from two different sources: watching battles and the tutoring program at this site. I was tutored by Black Leather Jacket, and he taught me how to predict by coaching me through situations until I understood how to respond effectively on my own. I battle a lot and that's basically the best way to approach the game, in my opinion. Just play the game. Losing doesn't matter if you're learning.

 

[obi]

If you build a "perfect team by the numbers" and it turns out horrible, then this doesn't mean that "statistics" are useless, but rather, that you used the wrong numbers.

"Instincts" being superior to anything is garbage. Saying you went by your instincts is just another way of saying that you are unaware of the processes by which you came to your conclusions. That doesn't mean these processes don't exist! The human mind is great at finding patterns, but rather poor at metacognition with regard to these patterns.

 

[Amelia]

Now, i won't say that i am a good or skilled player, but i consider myself average. And let me say this, my teams always sucks. I don't even pass the first point, knowing the stats of Pokemon.

I rely a lot on team themes, insticts, and predictions. Instead of knowing which Pokemon counters which Pokemon, i would rather use a theme, so that i don't need to counter anything at all, if it goes through. This has helped me a lot in many different aspects (for one, i'm not complaning about Garchomp at all, even though my team is quite weak against it.)

Also, i normally predict very well, which helps out a lot as well. I take a lot of risks, with the knowledge that if i get one correct, i'll control the game. Let's say that A has one counter against Weavile. If i could take a risk that would ensure that the counter dies, i would take it, because i know it would mean that my Weavile would fly through his team, instead of waiting for another opportunity.

Simply put, out of the first 6 steps, I'm not even used to the very first step.

 

[animenagai]

i agree with obi. 'instincts' are just either a glorified version of luck or another way of saying 'quick decision making'.

i'll talk about prediction here. many people say that prediction can just come down to luck and though that's not completely false it is also often incorrect. there are two parts to predicting correctly, knowing your opponent's style and knowing their team.

in terms of style, some people like to switch to pokes that can just 'take the attack' and thinking about what it'll do in return later. some people like sacrificing pokes more than others, and vice versa some like to save pokes with 1 HP as death fodder even if there's more risk involved. if you're playing a noob, just make the obvious move, they won't know what's coming so don't waste a turn. once you figured out the opponent's skill level and style, you can make good predictions.

the other part to prediction is knowing your opponent's teams. shoddy has the wonderful feature of telling you want pokes have already been shown and which ones have not. from that, if you are a decent team builder you can predict what's coming before you see their unrevealed poke. eg. if they have bliss, skarm, heatran and tentacruel, i would expect them to have celebi as it's probably a stall team and it works well with heatran. knowing your opponent's team also makes it easier to figure out what they need to keep and what they don't. hence you can predict when they'll make a sacrifice or an unexpected switch much easier.

this isn't some art i've mastered, but i hope that was at least somewhat useful :P

 

[Chou Toshio]

I agree with Obi, not in that "instinct" as a concept is bad, but that it is always better to be aware of what you're thinking than not. But, as Obi also said, realizing why you come to a conclusion can be difficult. Still, even if one is to consider this in retrospect after the game, one can improve.

 

[imperfectluck]

Wall of text ahead: read at your own risk!

1. Scouting: Ok, my Weavile's Ice Punch just did 30% to Swampert on the switch in. With this information, it's perfectly feasible to begin "approximating" the EVs on my opponent's Swampert, and formulating an idea of my opponent's team structure if he decides to send in Swampert against my Weavile early game. Take notes of your opponent's items, whether a Pokemon has Leftovers or not, what moves they used, and so on. If you can't remember, Notepad is a good idea.

2. Prediction and Risk Management: I prefer leaving mind games up to after knowing my opponent's playing style. Let's say I bring in Substitute Gengar on my opponent's weakened 40% Suicune, and my opponent decides to switch it out. In that case, knowing that he is a slightly more "conservative" player, I would be able to freely Substitute the next time this scenario occurs once more and cause serious damage with a Substitute up. Basically, know thy opponent.

3. Countering: As has been said, in DP it is "impossible to counter everything." Therefore, theorymon being put aside, just react to each and every situation as is, individually. Find the best possible scenario for each game as it plays out and always have a backup plan for winning. In order to win at the game of "countering" your team must acheive its objective (CM Raikou, Spikes, SD Yache Garchomp yada yada) for winning while keeping your opponent from putting his strategy together as efficiently/fast as you. Especially with offense vs. offense, a single "free turn" can mean everything (Intimidate making your opponent wanting to switch out, Choice locked moves, etc. etc.)

4. Instinct: Let's say that I have Deoxys-E remaining as my last Pokemon, just freshly sent out against my opponent's Salamence. My opponent has Salamence and Tyranitar remaining, both at full health. Assume Ice Beam OHKOs Salamence, Superpower OHKOs Tyranitar without any -Atk stages, and both Salamence and Tyranitar will OHKO Deoxys-E. In this case, if Deoxys-E Ice Beams and my opponent switches to Tyranitar to take a weak Ice Beam, then switches back Salamence into whatever to Intimidate Deoxys-E, I will lose to Tyranitar. I will only win if 1: I predict his Tyranitar switch with Superpower correctly on the first turn, or 2: I predict him predicting Superpower and Ice Beam his Salamence that stayed in the first turn. Ice Beaming into a Tyranitar switch or Superpowering a Salamence that stayed in mean certain loss (barring criticals.) With 4 possible scenarios, 2 leading to my win, and 2 to my loss, a 50% mathematical problem at face value. What statistics could I possibly rely on to ensure that my opponent is maneuvred into the move I want him to make? Is it "prediction" that I get it right? Luck? Skill? I could try this experiment with 1000 different people thinking out my move, and then try it again with 1000 people by rolling a die to determine my move. Would this really affect the outcome? I propose that "prediction" is a farce, as is "instinct" as there is no way to truly predict your opponent's move no matter what information you have. The only element is "luck."

5. Gimmick sets: I don't care if everyone else laughs at me for using Sunkern. I have a 90% win ratio with Sunkern on my team against good OU players, who are you to tell me it sucks? Ok, so that example was overdramatized, but if something can win, even if only by the element of surprise in that one game, it won the one game it needed to. Gimmicks are gimmicks because they only work once, as opposed to "reliable standard" sets that work whether or not your opponent sees it coming. "ChainChomp" has the potential to easily take out an opponent's physical wall such as Gliscor or Hippowdon with ease, but does not see as much play as SD Yache Garchomp because SD Yache Garchomp does its job much more consistently than can ChainChomp. Once you know your opponent is carrying a ChainChomp, it becomes that much easier to stop than if it was carrying its normal set (this applies to all Pokemon, not just Garchomp.) As has been said before: a set like Natural Gift Ice / HP Ice Heracross has lost it's effectiveness in the long run if it is forced to give up a moveslot for a single Pokemon (Gliscor.) However, in the short run, for that one game, it can pay off. I don't care if my HP Bug Swampert sees no use for 20 games in a row without seeing a Celebi, the 21st player I play might use Celebi and HP Bug will definitely pay off then and there, and I can certainly win with even that one moveslot restricted. If it wins the one game I would have lost otherwise, a gimmick has done its job. Gimmicks are not to be viewed as long-term solutions, in other words.

6: Team Building, Synergy: Synergy, to me, with regards to Pokemon, has to be an entire team goal, rather than clumps of Pokemon that work "synergetically," for example like "CeleTran + TarChomp." Celebi and Heatran are known to cover each other's weaknesses, Tyranitar can set up Sandstorm for Garchomp and both have astronomical physical offense scores, sure they can work well in pairs. Maybe using such a team has a gaping Mamoswine weakness so then you think to add Bronzong to your team. You aren't making your choices, your choices are being made for you. Synergy is when every Pokemon can set up the next for your own strategy. U-Turn Zapdos to lure out special walls, and to provide an "anchor" to center around, combined with potent physical threats that each can threaten to win at any time while supporting each other is one such team. For instance, combine Zapdos with Heracross, Tyranitar, Garchomp, Gyarados, and Metagross (Ok I know it sounds egotistical but bear with me for using my own team as an example of what I think is synergetic please.) Each Pokemon (other than Zapdos) is a "threat" with a high attack score that can sweep a team at any given time under the "right conditions." Therefore, I will endeavor to create the most beneficial conditions possible, such as Pursuiting anything dark-weak (Starmie, Celebi, Cresselia do NOT enjoy Tyranitar Pursuits therefore Gyarados can set up blahblah), sending out Heracross on Blissey for a chance to Swords Dance, Metagross can Explode at any given time to remove one of my opponent's physical walls, and just generally keep high pressure at all times with a plethora of Pokemon that have high attack stats. To reiterate, all of the 6 Pokemon on your team are truly a "team," rather than just "small groups working together."

7. Playing style: Pokemon is a game involving adaptability, where in each individual game you are faced with (at the outset, 9 possible moves per turn) involving 4 attacks and 5 switches. Very rarely, a team could possibly counter one so much that it becomes very difficult to win assuming equal skill level (a team of 6 Snorlax will likely never beat a team of 6 Lucario,) however, unless one person had prior knowledge of the other's team, that would be extremely rare. Pokemon is a game involving two humans, and humans are very prone to error, misjudgment, and simply making mistakes in general. In my opinion, whoever can better adapt to the situation facing him at the moment is the better player at that point in time.

 

[Taylor]

My opinion on "synergy" in Pokemon is an aspect of the game (mostly a combination of Pokemon) that is well-diversed with impressive results. It's relations between, for example, two Pokemon that are connected synergetically well.

For example, effective combinations as such; Celebi and Heatran / Tyranitar and Garchomp. These two pairs when coupled together are synergetically positive, thus they create an aurora of good synergy. They don't have to cover each others' weaknesses to have great synergy, but there must be something in the strategy between the two that is successfully benificial to the user. I don't believe that synergy, on its own, must consist of a well-balanced team where the whole squad is synergetically connected - a team as a whole that is synergized well is just impressive, overwhelming synergy.

We don't only have established, popular combinations of Pokemon that represent good synergy, as there are other aspects of the game where "good synergy" fits in particulary well. Movesets can synergize well - for instance, a Forretress and Starmie in the same team would synergize well so long as their movesets don't clash - let Starmie use Rapid Spin whilst your Forretress sets up Toxic Spikes and Stealth Rock, for example.

Unfortunately I don't have enough time to voice my opinion on all of the subjects before me.

 

[Cathy]

Another way to answer this question is "Is there a Pokemon AI that can beat every other team"? (Colin has claimed that he has created one, but we never saw it in action so w/e)

This isn't quite what I've claimed. I've just claimed it's trivial to obtain the best mixed strategy for the given turn if you know the opponent's team (so at a mid point in the battle). Of course, trivial doesn't mean practical; my implementation is so insanely slow that it can't even do three ply any time this year. My implementation is incredibly shoddy though; it could be better.

 

[chaos 9]

Colin, is that implementation by any chance that Anthony J whatever guy that everyone got their record reset if they battled him? I would be interested to battle an "unbeatable AI" just to see the team and the "thought" patterns.

 

[Battlerek]

i'm a tad skeptical about the whole instincts aspect. From experience i really dont need to use my gut feeling in battle, as calculating your opponent's strategy is vital, though overcalculation can sometimes be your downfall. Take this theoretical take on trick room teams, for example. After trick room your opponent may throw in a slow but powerful pokemon (say, metagross) and he takes out your trick room pokemon. instead of throwing in a slow pokemon to take your foe out, you bring out a weavile with suckerpunch. Granted weavile wont last long with trick room, but your assumptions of your foe's team has been broken. As such, you may resort to rescouting the team, wasting precious time and resources(pp).

 

[Aeroblacktyl]

Weavile doesn't get Sucker Punch.

 

[garo]

excelent thread tangerine!!, talking about the learning curve in a game like pokemon is very important because it is a game so deep that no one know everything about it, even the most experimented players and battlers learn new stuff, so when you are new at pokemon or simply like kingler dude says we face a generation change, what is the best way to learn??
first of all we battle and play pokemon because we love it right???, new players that starts playing pokemon because have heard that is a good game or whatever the reason and doesnt even imagine how deep and complex is the game, these players starts knowing nothing about mewtwo for example, they doesnt know he is a psychic type with advantages and disadvantages, they doesnt know how high is 154 points in special attack and why he is a legendary uber.
I will put my story as example, i know about pokemon since third grade of elementary school when a friend showed me a blastoise picture and told me "it is a pokemon called blastoise and there is a show on TV", since then i saw a lot chapters of the show, many of the movies and played a game from every generation, then in third generation i heard about netbattle somewhere ( i think i read something in FAQS when i was looking for some info about emerald version) i downloaded it and started battling with the team i ever dreamed: venusaur, charizard, blastoise, gengar, dragonite and mewtwo, they were my favorites and i put only attack moves on them lol, obviously i didnt win a battle (maybe one but with god´s help lol), i was very disappointed, but anyway i started learning a lot of things like why not to put three fire moves in a poke and why special attacks barely scratchs blissey lol, and in no time i become a good player and won battles, i spent hours reading about pokemon abilities, moves, breeeding, EVs, etc. etc. and i read it with joy because i like pokemon (indeed i think i remember more things about pokemon that about the stuff i study at school lololol).
so in conclusion i think that with experience you learn faster, with battling you learn what is a counter and how to use it, how to predict the opponent, why a moveset is good on that pokemon and not that that is good just because someone told you that. of course that reading, doing math and statistics is essential but you will do it and experience it with battles the faster you can. i think this is the shortest way to overcome the learning curve.

 

[Bam]

I just realized that I never got around to posting in this thread, so I figured I'd do it.

Obviously I haven't covered everything - but I feel as if this has enough discussion worthy content... for now. I will update the OP as needed.

To me, the biggest part of getting started on the learning curve is, other than memorizing the info and data, getting over the concept of roles. It is actually pretty similar to what you and others have stated about getting over the idea of building teams with countering in mind (which you actually pointed out today in my latest RMT, thanks for the help btw), and may even be classified in the same category (Blissey = Starmie counter = special wall and back and forth). *it has also been discussed in your other role thread, but I thought I'd add it here*

For me, the shift from trying to counter as much as possible to working on a "game plan" while still covering the bigger threats occurred when everybody started sigging that quote of MoP's, which in effect said "why worry about countering everything when it's impossible - worry about setting up your own strategy", and I payed attention to it.

Regarding the last topic in the OP, I seriously doubt that there will ever be a perfect (or the best) team. With the tier lists being in a constant state of change, people complaining about x needing to be uber or y needing to be moved down, and newer seem-gimmicky but work sets being posted every once in a while, the metagame will never even settle enough into a position where a perfect team can be designed. Besides the fact that countering everything is impossible, luck screws over said team, and when a key Pokemon (or any Pokemon, as the perfect team would require every Pokemon to play a huge role) is taken out, that opens a whole for the other team to go through, similar to the o-line making a hole for the tailback to get through.