1. New to the forums? Check out our Mentorship Program!
    Our mentors will answer your questions and help you become a part of the community!
  2. Welcome to Smogon Forums! Please take a minute to read the rules.

Risk/Reward Analysis

Discussion in 'Locked / Outdated Analyses' started by Nate, Aug 8, 2006.

  1. Nate

    Nate
    is a Forum Moderator Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Dec 20, 2004
    Messages:
    853
    A concept often overlooked while discussing prediction is the concept of measuring the risk and reward of a move. This article will look at the basics of measuring the risks and rewards of moves, strategies for minimizing risk or maximizing reward, and pokemon who skew the risk/reward relationship.

    Risk/Reward in Basic Scenarios

    There are two types of ways in which risk can be measured. You can measure risk in what your opponent is most likely to do, or you can measure risk by the move your opponent could use which would hurt your team or overall strategy the most. Both of these applications have scenarios in which they can be applied, and differentiating between the scenarios is key as well.

    Scenario 1:

    Player 1: Regice & Suicune
    Vs.
    Player 2: Zapdos & Metagross

    In many cases, you can have an extremely good idea of what your opponent will do before he does it. If a Metagross comes in on a Regice, you have a pretty good idea that it will use Meteor Mash. Before moving, you unconsciously take the risk/reward into account. If you stay in and he uses Meteor Mash, Regice is dead. That is a big risk. However, if you switch to Suicune and he uses meteor mash, little damage is done. That is not very risky whatsoever. If you are playing conservatively (in this case intelligently) you will take the switch to Suicune. The analysis gets far more interesting from the other end, playing as Player 2. You have just brought in a Metgross on a Regice. You have a Zapdos on your team as well. You can be fairly certain that your opponent will switch his Regice to his Suicune. Let’s analyze the situation. You can be almost certain that your opponent will switch to Suicune, as it would be a great risk for him to keep Regice in. This is an example of the first way in which to measure risk. Your opponents move choice is almost definite, so you can use his lack of options to switch to Zapdos.

    This prediction is not cyclical because of the lack of risk for Player 1. He is forced by his opponent to switch to Suicune, but his opponent possibly predicts that and switches to Zapdos. He does not Ice Beam with Regice, however, because there is no risk in his switch to Suicune. Regice can continually wall Zapdos, and Suicune can continually wall Metagross. Player 1 does not have to worry about prediction games at this point, because there is very little risk to playing his opponent at face value.

    Scenario 2:

    Player 1: Medicham @ Choice Band & Gengar (low hp)
    Vs.
    Player 2: Regirock (low hp) & Salamence @ (fairly high hp, dancer)

    Medicham has come in on Regirock. As player two, this scenario should be analyzed in the second method: considering the most dangerous outcome. You are not sure whether the opponent’s Medicham will use Hi Jump Kick or Rock Slide. Rock Slide will kill Salamence, and Hi Jump Kick will kill Regirock. If you switch to Salamence as your opponent Hi Jump Kicks, you will survive and can sweep his team after just one dragon dance. If your opponent uses Rock Slide as you switch to Salamence, you lose. If your opponent switches to Gengar as you switch to Salamence, you also lose. There are two worst-case scenarios in this situation: you take a Rock Slide from Medicham or your opponent switches to Gengar as you switch to Medicham. Both of these losses stem from making the switch. The reward is greatest if you take the small risk that Medicham Hi Jump Kicks (resulting in a Regirock death). If you Rock Slide from Regirock, you force the win no matter what your opponent chooses.

    This scenario is very specific, but is chosen from an actual battle as it exemplifies how risk/reward can be far more important than prediction. And it is also a very nice transition into the next section of this article.

    Pokemon is Chess: Playing Transparently, Conservatively, and Riskily

    Playing conservatively is most feasible when your team has an advantage over your opponent. In this style of play, you take care to minimize risk at every single turn. Say your opponent has no Jirachi counter, a far more common instance than I would have thought. If your team has few weaknesses to the opposing team, you should play extremely conservatively. If you play with a team with three or four walls or tanks, this should be no issue. Simply switch in a pokemon that can completely stall whatever pokemon they switch in. Only one prediction is needed in the entire match. After your opponent is struggling to get through your tanks, a switch to Jirachi on a Pokemon such as Blissey or Milotic will swing the game permanently in your favor. This strategy can be quite boring, but quite effective when you know you have a team advantage over your opponent.

    In every battle, try to find the one pokemon that the opposing team has the most trouble countering. If your team has an advantage over your opponents, then move cautiously to minimize risk until you are certain you can get that pokemon in. Before every turn, carefully consider both the most likely move of your opponent and the worst-case scenario. If your team is matching up well, however, then worst-case scenarios will be largely irrelevant. If your team matches up poorly against your opponent, you will have to ignore worst-case scenarios completely as you struggle to get your most-damaging pokemon in. The worst-case scenario is hardly worse than the average case, in which you are worn down or beaten by a pokemon for which you have no counters. By manipulating the most likely move of your opponent, you can create opportunities for your most damaging pokemon to switch in and attempt to equalize the match.

    Dugtrio, Magneton, and other Skewers

    Certain pokemon have an huge impact on risk/reward analysis. Dugtrio is possibly the greatest of these. If you are using a Celebi, you may switch in and out a few times to see if your opponent has a Dugtrio. Once you are aware your opponent has a Dugtrio, your risk/reward analyses are skewed for the rest of the game. Dugtrio forces you to consider worst-case scenario options continually when thinking about switching in Celebi (worst-case being they switch to dugtrio at the same time). This can cripple your defense against pokemon you may depend on Celebi to counter, such as Swampert. Dugtrio’s presence, even without usage, forces you to play with far greater risk no matter the strategy you choose. If you choose to go to Celebi, you are taking the large risk that Dugtrio switches in, for the small reward of forcing a Swampert switch. If you choose not to switch to Celebi, you are not using your usual minimal risk defensive strategy for that mud monster.

    Magneton operates much the same way. Oftentimes people fish for Magnetons by Roaring/Whirlwinding unnecessarily with their skarmories. I encourage you to put magneton in to get roared, as that will greatly enhance the usability of physical tanks. The knowledge that Magneton is waiting in the wings will greatly decrease your opponents’ urge to block your Curselax with Skarmory. They will be forced to, at the minimum, alternate physical walls in order to keep you guessing when they will bring their Skarmory out. Unlike Dugtrio, however, Magneton has a very limited usage. He is extremely useful against Skarmory, obviously, and to a limited extent against pokemon such as Gengar (presuming you can switch it in while Gengar does not use Substitute). Using Magneton is another bet, seeing as it is riskier to use than zapdos, but the potential rewards are far greater. The great effectiveness of both Dugtrio and Magneton lies in creating a sense of enhanced risk in the opponent. It greatly increases the risk of using specific very common pokemon, and that opens up the defenses of teams.

    A pokemon doesn’t need to have a trapping ability in order to have a great impact on an opponent’s sense of risk. Choice Banders are easily the greatest example of this, especially the ridiculously powerful ones such as Slaking, Medicham, Metagross, or Salamence. Pokemon such as these, with no completely safe counters, force your opponent into situations where they have no opportunities to minimize risk. The best they can do is guess at what move you have chosen, and choose an appropriate counter. Prediction is key in using these, but even guessing right only occasionally will have an extremely large impact on the game. These pokemon with just a few hits open up the opposing defenses for previously walled pokemon. Versatility and surprise can be an important factor in risk analyses towards the beginning of the game. A Choice Band Gengar lures a celebi switch, only to do huge damage with a stab-ed Shadow Ball coming from a pumped attack stat. Surprise lets the opponent think they are minimizing risk, only to be placed in a handicap. After the surprise is gone, however, the surprising pokemon (such as choice band gengar) is often placed at a disadvantage to its lack of power and now lack of versatility. Thus it is often advisable to fake a normal set until such a time when a surprise can open up the game.

    In Closing

    The most important analysis to be done is to compare teams. First, figure out which team holds the advantage. Second, figure out which pokemon on both teams can do the most damage or change the course of play. If your team has the advantage, play conservatively while minimizing opportunities for your opponent’s game changer to switch in. Play conservatively until you get a sure-fire opportunity to switch in your game changer. If your opponent’s team has the advantage, you will need to use surprise, predictions, and most likely a bit of randomness in order to regain some momentum. By making yourself unpredictable, you can attempt to force a switch which allows your game-changer to enter play.
    3 people like this.
  2. giga

    giga
    is a Battle Server Moderator Alumnus

    Joined:
    Jun 15, 2005
    Messages:
    3,350
    Great writing, this is should be a good source of info and how to play mind games/whatever you call it with your opponents. Nice :D
  3. husk

    husk
    is a Tournament Director Alumnusis a Site Staff Alumnusis a Forum Moderator Alumnusis a Battle Server Moderator Alumnusis a Past SPL + WCoP Winner

    Joined:
    Jan 9, 2005
    Messages:
    2,067
    I agree with everthing here. I think a section on the usefulness of sacrificing pokemon should be added as that is a big part of risk vs reward (and if your opponent will sack a certain pokemon at a certain time or not - which is much harder than sacrificing stuff yourself imo).
  4. thecrazylazyguy

    thecrazylazyguy

    Joined:
    Apr 13, 2006
    Messages:
    628
    I like this since I read basics thread on risk reward and how it all looped back together but it did not seem quite complete. I think this should be added since its something many newer players donttake into consideration
  5. Emerald_Ryuu

    Emerald_Ryuu

    Joined:
    Apr 27, 2006
    Messages:
    202
    This topic should be stickied; these are excellent points to keep in mind while battling.
  6. gene

    gene GENE
    is a Tournament Director Alumnusis a Site Staff Alumnusis a Smogon IRC AOp Alumnusis a Forum Moderator Alumnusis a Battle Server Moderator Alumnusis a Past SPL + WCoP Winner

    Joined:
    May 1, 2005
    Messages:
    2,723
    I might have missed it somewhere, but a section on Exploding could help some people. It is always a risk/reward situation when you are facing off Blissey v Gengar or in a similar situation. You have to think about: if Gengar is needed badly on a team, so badly that they won't dare explode, or if they really need to get rid of Blissey badly. It is one of the most major parts of this whole thing imo.
  7. Warthog

    Warthog
    is a Forum Moderator Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Dec 20, 2004
    Messages:
    3,967
    I love the article but I have one gripe. I dont see Pokemon as Chess but as Poker. Chess takes skill while Poker takes more risk and luck. In my mind's eye, its either "All in" or "Im out". Just like in Poker, you must make a decision between the two choices.
  8. Nate

    Nate
    is a Forum Moderator Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Dec 20, 2004
    Messages:
    853
    I can definetely sympathize with you on that one. That section started out as Pokemon is Chess: Playing Conservatively and was only about taking advantage of a perceieved advantage. I enjoy playing pokemon as if it was chess when I know I have the better team; no prediction, just forcing the opponent slowly into defeat no matter what they do. Then I realized it was kind of dull so I expanded it. The section title doesn't make quite as much sense anymore though, I suppose.
    1 person likes this.
  9. david stone

    david stone Fast-moving, smart, sexy and alarming.
    is a Site Staff Alumnusis a Smogon IRC AOp Alumnusis a Programmer Alumnusis a Super Moderator Alumnusis a Researcher Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Aug 3, 2005
    Messages:
    5,150
    I don't really have anything to add, well done. :)
  10. makiri

    makiri My vast and supreme will shall be done!
    is a Tiering Contributoris a Smogon IRC AOp Alumnusis a Forum Moderator Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnusis a Past WCoP Winneris a SPL Winner

    Joined:
    Jun 24, 2006
    Messages:
    4,233
    Well written, good job
  11. Aeroblacktyl

    Aeroblacktyl The pizza doesn't scream in the oven! LOL!
    is a Tutor Alumnusis a Tournament Director Alumnusis a Site Staff Alumnusis a Team Rater Alumnusis a Forum Moderator Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnusis a Past SPL + WCoP Winner

    Joined:
    Feb 19, 2005
    Messages:
    8,335
    IMO this doesn't work for everyone. That being noted, imo Intimidation of a player plays a huge important role on how you would battle someone. If someone you think is better than you playing agianst you, I'm pretty sure you wouldn't dare make a single high risk move.
  12. Giga Punch

    Giga Punch "No I'm not a porn star but I eat somebody"
    is the Smogon Tour Season 5 Champion

    Joined:
    Nov 18, 2005
    Messages:
    758
    Good Job

    Agreeing with MoP somewhat..
  13. Jumpman16

    Jumpman16 np: Michael Jackson - "Mon in the Mirror" (DW mix)
    is a Smogon IRC SOPis a Site Staff Alumnusis a Team Rater Alumnusis a Battle Server Admin Alumnusis a Researcher Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Dec 19, 2004
    Messages:
    4,768
    I don't agree with this because if there's high reward, like trapping their 35% HP Milotic with Dugtrio as it tries to switch in to get a Recover that would spell eventual defeat at the hands of an opponent who outclasses you, you won't necessarily play it safe, because what you do have to lose? You're the underdog anyway, so why not go all-out and take some risks if it can really swing the battle in your favor quick?

    The best "real life" example of this is poker tournament where amateurs find themselves at the table, late in the tourney, with a few pros who they know have them outclassed. You see these amateurs pushing all-in preflop much, much more, not giving the pros a chance to pick them off one pot at a time by outtplaying them after the flop. It happens all the time in poker, and there's no reason to believe that this doesn't at least somewhat translate to pokemon.
  14. FiveKRunner

    FiveKRunner

    Joined:
    May 27, 2006
    Messages:
    812
    Wow, this is a great guide. One of those things that I'm sure will help people, especially when they're just starting out. Some of those little things that can really turn a battle in your favor.

    This definitely needs to be added to the articles.
  15. GaviaO

    GaviaO

    Joined:
    Aug 27, 2005
    Messages:
    63
    Very nice article, Nate!! I try to always take this risk/return (in-game lol) in consideration...
  16. Undertaker

    Undertaker

    Joined:
    Jun 1, 2006
    Messages:
    164
    Yeah I agree with Warthog; chess requires you play conservatively, but as a few great players have shown, you must also take risks. But in chess, cold hard calculation (skill) beats risk most of the time, while poker requires a delicate balance of luck and skill to win.
  17. gbagcn

    gbagcn

    Joined:
    Sep 11, 2005
    Messages:
    52
    This article needs a section about challenge cup. Even luck plays a large role in challenge cup there are still a few things you can do to make yourself win more often. Here are some examples:

    1. Since many pokemon in challenge cup only have normal type moves make sure you make an extra effort to keep your ghost, rock, and steel type pokemon alive as long as possible.

    2. If you notice a few of your pokemon have much better movesets than the rest try to avoid sending them out until you have more information about your opponents team.

    3. Sending out your weak pokemon is a good idea if you are unsure what types of moves your opponents pokemon has. Once you know the moves of your opponents pokemon you can better counter them.

    4. Make sure to keep pokemon with water attacks alive as long as possible. This is because water attacks are super effective against rock types and do normal damage to steel and ghost types. Since these types of pokemon are the most important types in cc having a move that does a good amount of damage to them is very important.

    5. Try to predict type of attack your opponent will use based on his switching behavior. For example if you have a rock type out and your opponent switches to a normal type that means that normal type most likely has a move that is super effective against your rock type.

    6. Try to trick your opponent into thinking you have a move type that you really don't. For example if your opponent sends out a rock type early in the game if you don't have many pokemon with an attack that does normal damage or better to it send out a water pokemon that has no normal type attacks. If you send out a water type pokemon your opponent most likely thinks it has a water type move since the chances of this being true are normally high. Because of this your opponent will most likely switch.

    7. Avoid using your best moves and best pokemon early in the game even when it would normally be a good idea to do so. This way your opponent has less information about your team so he can't make informed decisions on how to counter it.

    8. Late in the game if you forgot what moves some of your opponents pokemon have check the battle log to see what moves they have been recorded using.
  18. Amazing Ampharos

    Amazing Ampharos
    is a Researcher Alumnus

    Joined:
    Jul 11, 2005
    Messages:
    1,559
    I find the secret to winning as many of these as possible is to ask myself what the opponent would do. For instance, in the Regice vs Metagross, I very well may thunderwave the incoming Zapdos which would be a win for me. It has the added advantage of winning if meteor mash is used and misses or if he's running a thunderpunch Metagross and tries to predict on my Suicune. Being unexpected at the right time will almost always win a Pokemon match, more so than playing safely. Also, when up Pokemon, taking risks is still important(remember, now you can afford to lose), but the key is to see which risks are good to take. Risks that you allow stat boosting or set-up are almost always unacceptable; risks that open their team up to a sweep while standing to lose a semi-useful Pokemon for you are often the secret to preventing comebacks. Many people who are in positions like 6-4 will try to keep up those two Pokemon when they are often better served taking a chance to drop to 5-4 with the prospect of destroying the enemy's last defense. The longer you let it drag on while winning the more likely they are to have a trick up their sleeves. Of course, the fundamental rule is to not go for a 6-0 win if a 5-0 win is safer, and if you have a Pokemon they can no longer counter it is almost never worth taking a risk with it. If I had a metagross that could sweep them and was up a lot, I'd sacrifice a Pokemon to bring it in instead of risk getting frozen by a predicted ice beam.
  19. Mekkah

    Mekkah
    is a Super Moderatoris a Tutor Alumnusis a Site Staff Alumnusis a Smogon IRC SOp Alumnusis a CAP Contributor Alumnusis a Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
    Super Moderator

    Joined:
    Feb 8, 2005
    Messages:
    14,041
    Good read, seems to float along with my long-term thinking analysis.
  20. Hot Nikkels

    Hot Nikkels

    Joined:
    Jun 30, 2005
    Messages:
    211
    Good thinking making a thread like this, I've noticed alot of players tend to think in absolutes and in black and white too much when making predictions. One example of this that I use is to keep a poke that you know will be killed by Spikes as soon as you send it out. Save these guys for when you're "behind" a turn to even things out, such as when you're losing in the late game and you know your Heracross or Raikou or other sweepertype can't take both Spikes and an incoming attack.
  21. Basics

    Basics

    Joined:
    Mar 10, 2005
    Messages:
    3,725
    I think the most important thing to stress in terms of risk/reward in Pokemon is how, unlike in more traditional applications of risk/reward, such as fighting games, the concept in Pokemon is much more abstract and subjective. That is to say, in fighting games, your risk/reward is very simple. How much damage will this do if it hits? How much damage can they do to me if it misses? And that's pretty much it, although obviously other things such as knockdowns, positioning and resets come into play.

    In Pokemon, however, it's very much different. Damage isn't all that matters, since there's also setting up your other Pokemon (OK, so I have Gengar out and I think he's switching in Snorlax. I could Sludge Bomb for some okay damage, or I could switch to Medicham and set up some good mind-games). You also have the luck factor to consider (OK, I have Snorlax out and I think he's switching to Salamence. I could switch to Milotic and be ready for it, or I could Body Slam and hope to Paralyze him.)

    This isn't that thorough, but I think you get what I'm trying to say. The abstract aspect of risk/reward is the only thing that really adds depth to the game of Pokemon.
  22. Hot Nikkels

    Hot Nikkels

    Joined:
    Jun 30, 2005
    Messages:
    211
    You made your point but I think that applies to fighting games just as well. In most games I would rather go for the good meaty wakeup/guard break/crossup than straight up damage unless it's late game and makes all the difference. Take CvS2 C-Ken for example, whenever I hit a cr. shortx2 into lvl2 Shoryuu Reppa and cancel into command roll, I'd usually rather go for the crossup HP Shoryuken than lvl1 Shinryuuken because of the wakeup opportunities it offers. Well anyway, I do agree with what you're saying I just like talking about fighters with people. By the way Basics do you ever play at Orbit?
  23. Basics

    Basics

    Joined:
    Mar 10, 2005
    Messages:
    3,725
    Fuck yes I play at Orbit, I get murdered by the competition there, though. :( Or at least most of the time. My Ryu is really coming along, though. CvS2 I just play A-Balrog, Rolento, Eagle. I can't even do all their CC's reliably, I just fuck around.

    But no, what I mean is, you have different things to consider. For fighting games, it's all pretty black and white. Pokemon's choices are alot more severe, and impacting. If you make a bad decision and switch your Skarmory into Metgaross' Thunderpunch, your Skarmory takes a bit of damage. So what? Well that means that your Skarmory can't switch into my Salamence safely anymore since Fire Blast is an OHKO.

    In a fighting game, during early footsies, if you score a knockdown with c.hk, then wake up with c.lkx2, I block it and you try to walk in for a throw and I dash back, we're back to normal. While, comparatively, these situations are similar, the slippery slope aspect of Pokemon makes your decisions matter alot more. One fuck up, even a minor fuck up can cost you the game. At least in ADV anyway.
  24. Hot Nikkels

    Hot Nikkels

    Joined:
    Jun 30, 2005
    Messages:
    211
    Yea I see what you mean, but then there's Marvel..
    Last time I went up to Orbit I remember C Royd beating me with one hand. With his girl sitting on his lap. Damn. haha I did pretty well in SC2 though for what ti's worth.
  25. jpchato

    jpchato

    Joined:
    Mar 27, 2005
    Messages:
    1,170
    Critical hits aren't random in fighting games either. When I did play, I always tried to think at least two steps ahead, while minimizing risk. Since there isn't a block button in this game, just switch a lot, but not obviously. If that makes any sense. Luck plays such a huge factor in this game that you REALLY REALLY REALLY want to take the smallest risks possible.

Users Viewing Thread (Users: 0, Guests: 0)