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.

Attacking Type Effectiveness

Discussion in 'Pokémetrics' started by X-Act, Jan 22, 2008.

  1. dirtybirdy

    dirtybirdy

    Joined:
    Nov 21, 2007
    Messages:
    167
    I'm feeling the same way to be honest. Take ice for example, theres not much point to knowing physical ice is the best attack if all we have to work with is ice shard, avalanche and ice punch...I think.
  2. darkie

    darkie mfw i see alison brie
    is a member of the Site Staffis a Smogon Social Media Contributoris a Smogon IRC AOPis a Super Moderatoris a CAP Contributor Alumnusis a Contributor Alumnusis a Smogon Media Contributor Alumnusis a Battle Server Moderator Alumnus
    Public Relations

    Joined:
    Dec 25, 2005
    Messages:
    6,128
    I meant, on the special side, Water and Ghost are ranked higher, while everything else is the same.
  3. AJC

    AJC

    Joined:
    Jun 7, 2005
    Messages:
    2,265
    i wonder how diffrent those special side number would be if blissey and chansey didn't exist
  4. Syberia

    Syberia [custom user title]
    is a Pre-Contributor

    Joined:
    Jun 22, 2005
    Messages:
    5,020
    Read my post directly above yours. As far as I can tell, it's useless in a competitive sense.
  5. DevilinDenial

    DevilinDenial

    Joined:
    Aug 7, 2006
    Messages:
    1,389
    Although you can always use it to add to your logic of best move to use when you know your opponent is about to switch. If this was out when D/P started I'm sure this would be more gamebreaking, but now it's just another reason for people to use Stone Edge when they bring out Heracross.
  6. Slice n Dice

    Slice n Dice

    Joined:
    Nov 3, 2007
    Messages:
    2,441
    Yea, Ice> Ground because not nearly as many things resist it, while still having an excellent 4 type coverage. Hits annoying as fuck stuff like Chomp, Mence, and Gliscor(aka Hi I'm Mamoswine coming to rape you're Sandstorm team!)

    Once again, great research X-Act.
  7. RaikouLover

    RaikouLover

    Joined:
    Aug 31, 2006
    Messages:
    2,430
    The math is good. But I personally think water is the most effective attacking type, especially for physical moves. Only Grass, water and dragon resist it, and those are limited in a certain team. More importantly, as far as physical attacks are concerned, most physical walls cannot wall water attacks at all, because of SE. And boosted water attacks (via rain dance, choice band, or Dragon Dance) hit even the steels for decent damage. Just to give you an idea, a lonely Kingdra holding mystic water will 2HKO forretress and skarm with SR... and a DD Gyarados will do even more damage with a Life Orb.
  8. Chou Toshio

    Chou Toshio @Fighting Necktie
    is a Forum Moderatoris a Community Contributoris an Artist Alumnusis a Contributor Alumnusis a Smogon Media Contributor Alumnusis a Battle Server Moderator Alumnus
    Moderator

    Joined:
    Aug 16, 2007
    Messages:
    8,212
    Wait, does this attack chart take into account what attacks are available to the pokemon? For instance, does it take into account that the only physical ice attacks we have are ice punch (rare) and avalanche?

    If not, than one could say that the metagame has developed in a way where pokemon like garchomp thrive because of the absence of attacks that they are weak too-- which affects their use, and therefore affects this chart. Not saying that's a bad thing, just pointing it out. If we had a wide-spread close combat for ice attacks, the dragons might see a lot less use than they do.
  9. Deck Knight

    Deck Knight A Knight for the Aegis
    is a Forum Moderatoris a CAP Contributoris a Smogon Media Contributor
    Moderator

    Joined:
    May 27, 2005
    Messages:
    7,549
    My problem with this is that it doesn't seem to account for the most important factor in attack effectiveness: Base Power.

    For instance, theoretically Ice is better as a physical attacking type.

    The problem is, Ice Punch is the strongest physical Ice Move, and as of you the only pokemon to carry it competitively is Weavile and Azumarill.

    Mamoswine utilizes Ice Shard/Ice Fang/Avalanche with a large amount of success, but outside of those two physical ice attacks are rare. In most situations, Blizzard is going to do more than Ice Punch to any given target. Ice Beam will to a much lesser extent, but still often enough.

    Rock is theoretically a good special attack, but nothing uses Power Gem and next to nothing uses HP Rock.

    Flying, Dark, Fighting, and Ghost are probably the most representative, given the relatively close base powers of their strongest physical and special moves.

    I know you use base power in the second formula, but personally I think it's kind of a pointless average. OU consists of both fragile sweepers and tanks and walls. The sweepers tend to take the average damage way above where it would be.

    Good work on this, though.
  10. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    Okay, trying to answer and comment on the loads of comments I received.

    Thick Fat, Flash Fire, Water Absorb, Wonder Guard, etc. were ALL accounted for. Practically all damage-lowering or negating abilities were accounted for. I also took into account Tyranitar's Special Defense boost. And yes, I think Ground is relatively down the list because of the many instances of Levitating and Flying Pokemon in the OU metagame.

    About not including base power: this is done on purpose! It also neglects attack stats on purpose. That way, you can actually compare which would be the better hitting attack from two specific Pokemon. For example, which hits harder on average, an Earthquake from 359 Attack Garchomp, or a Sky Uppercut from 372 Attack Blaziken? Let's answer that question:

    Garchomp: 359 x 100 x 1.5 / 88497 = 61%
    Blaziken: 372 x 85 x 1.5 / 75295 = 63%

    So Blaziken would hit slightly harder on average, even though 359 x 100 = 35900 is much larger than 372 x 85 = 31620. (Maybe because Blissey is the commonest Pokemon?)

    The above calculation is also an example to those people who can't understand what these numbers would be useful for. If you still can't see a use for them, then tough luck I guess... you can never please everyone. :)

    About Normal being so down the list: remember that many overused Pokemon resist Normal. Gengar, Tyranitar, Bronzong, Forretress, Skarmory, Dusknoir... those are all extremely overused Pokemon and all resist or are immune to Normal. Remember that those numbers are also reflective of the metagame, and that, when all is said and done, Normal cannot hit anything for supereffective at all.

    As someone else commented, Fire is the only type that hits better from the special side than from the physical side. Why is this? Because most overused Pokemon that are weak to Fire have lower SpD than Def. It's better to hit Skarmory with an HP Fire than with a Fire Punch.

    About the calculations sometimes going over 100%, how can I know if they do when the Attack and Base Power of the moves are not even considered? Hence I cannot really fix that. :)

    I need to make it clear again that these numbers reflect December's metagame as of December's Shoddybattle stats. They would have already changed slightly by now, since there has been almost three weeks of play, and a recent introduction of a Pokemon (Deoxys-S) in the metagame to compound matters. They can be updated every month as per the Shoddybattle stats, although, honestly, I don't expect too much of a change since usually the top 50 Pokemon or so are always the same. Apart from this Deoxys-S case, of course.
  11. TAY

    TAY You and I Know
    is a Team Rater Alumnusis a Super Moderator Alumnusis a Smogon Media Contributor Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Nov 7, 2007
    Messages:
    1,542
    To account for base power in the formula might make people think that brick break is a better option than ice punch in certain cases, simply because Close Combat would have brought the PATE of fighting moves way down (Head Smash would do the same for rock). Something like hyper beam would throw the whole thing off, since it is learned by most pokes and has huge Base Power--not to mention the fact that it's near impossible to numerically account for the recharge, or for recoil, or the effects of Outrage, etc.

    X-Act can correct me if I'm wrong, but I believe that the chart was meant to be used in combination with the formula to determine a certain move's effectiveness on a certain pokemon, or to help determine which pokemon to use over another based on its moveset/STAB's.

    The way it is currently set up--with the chart AND the formula--is a lot more useful in terms of application to more moves and application to more pokemon (and it was a significant amount less work, I'm sure).

    For example, say you're trying to decide between Night Slash and Brick Break for your Max-Attack Sandslash (an unlikely scenario to be sure, but it's just an example). For night slash, compute (378*70)/82175=32.2%. For Brick Break: (378*75)/75295=37.7%. And voila, brick break is a statistically superior choice. (since both are physical moves the attack value is irrelevant; I just wanted to get a meaningful percent. But keep in mind the attack values would be different if a physical move is compared to a special one.)

    I'm not saying that the chart and formula are infallible; Staraptor's Close Combat is rated below Double-Edge, which is obviously not the best choice. But with a bit of common sense, the chart can be useful for generating new testing ideas, picking filler moves, deciding which attack to use if much of an opponents team is unknown, &c; and besides all that it's a great referential resource.

    This is pretty awesome work. --TAY

    EDIT: As it turns out, X-Act beat my post by 3 minutes. Oh well.
  12. astrohawke

    astrohawke

    Joined:
    Jul 19, 2007
    Messages:
    587
    I feel that the actual % of average damage calculations has no real practical applications. The example of infernape averaging 95-112% with CB flare blitz convinced me of this.

    In a competitively battling sense there are far too many variables to be taken into consideration. Stuff like overkilling those that are 4x weak by a ton of damage can skew the outcome and make something look better than it really is. Also in an actual battle you are far more likely to come across the pokemon that will resist that attack. Overall flare blitz looks awesome but in a battle, it's far more likely that outrage will sweep the team than flare blitz purely because the chances of a fire attack being resisted are greater than that of a dragon attack being resisted on any single team.

    However, there is one application I see this useful for. Basically if you don't know what attack to use, using an attack type high up on the list will yield better results on average. For instance, at the start of the battle I have my gengar out against a celebi. It's almost guaranteed that they will switch out of a shadow ball to something that resists it so I decide to predict. But I have no idea what pokemon they have so I can't make any real predictions. Using the list, I can choose to use HP ice or even focus blast because they are so much higher on the list than shadow ball and have a greater chance of hitting the switch in for good damage.
  13. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    It has an application, astrohawke: to compare two attacks together.

    Say you have:

    Aerodactyl @ Choice Band
    Jolly
    42 HP, 252 Atk, 216 Spe

    Stone Edge
    Earthquake
    Ice Fang
    Fire Fang

    This is cut-and-paste from the Aerodactyl analysis. So we have Aerodactyl with 463 Attack.

    Stone Edge: 463 x 100 x 1.5 / 73926 = 94%
    Earthquake: 463 x 100 / 88497 = 52%
    Ice Fang: 463 x 65 / 70302 = 43%
    Fire Fang: 463 x 65 / 74089 = 41%

    That's the statistical damage they will do. Notice that Earthquake has 100 base power and Ice Fang only 65, yet Earthquake will only deal 9% more than Ice Fang on average, even though Earthquake has a whopping 54% more power! (base power for base power.)

    In short, you do not use this to say "hey Stone Edge from CBAero almost OHKOes every Pokemon in the game!", but to compare it with other attacks that you're using.
  14. astrohawke

    astrohawke

    Joined:
    Jul 19, 2007
    Messages:
    587
    The thing is, after knowing these percentages and being able to compare them, what do they actually do for us? So now I know that despite earthquake having a lot more BP than ice fang, ice fang still does similar average damage overall. What then can I do with this knowledge? What do I gain from comparing the damage? Because it certainly doesn't tell me that I will be able to nearly ohko every pokemon on my opponent's team with stone edge or that I will do similar damage to all my opponent's pokemon with earthquake and ice fang.

    I still feel that the raw values can serve to tell me which attack types will be the most effective on average against most OU pokemon in the metagame. The percentages however are really just arbitrary values used to "compare" attacks. They are derived from calculations involving all pokemon and at the end of the day serve no purpose in an actual battle with only 6 pokemon in a team.

    Unless I'm still not understanding something important here, in which case I'd like a more detailed explanation of its uses other than to "compare attacks". Essentially, how does "comparing" the attacks help me practically?
  15. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    My point is that it's useless to know that Ice is the best type when your best Ice Attack has only 65 Base Power. It's useless to learn Ice Beam (which would equate to the best attack in the game), when your Special Attack stat is low (ask Delcatty). Hence, the attack stat and base power of the move are important.
  16. xcfrisco

    xcfrisco

    Joined:
    Mar 17, 2006
    Messages:
    595
    This topic really interested me on how effective are these attacks in OU. Instead of posting my own thread, I think these findings go pretty well with what X-Act posted.

    There are 49 pokemon in OU. This list shows how many of these pokemon are hit for Super-Effective, Neutral, and Resist, or completely Immune damage. I ignore 4x damage and resist for neatness sake. I used the types which I felt were the most common moves used in the OU metagame (The top five in X-Act's post plus ground moves). I took the abilties of each pokemon to account as well (Thick Fat, Levitate, and Flash Fire.)

    Rock-type move
    SE-9
    Neu-27
    Res-13

    Flying
    SE-9
    Neu-30
    Res-10

    Ground
    SE-10
    Neu-21
    Res-4
    Imm-14

    Fight
    SE-11
    Neu-20
    Res-15
    Imm-3

    Ice
    SE-14
    Neu-19
    Res-16

    Fire
    SE-16
    Neu-19
    Res-13
    Imm-1


    There are a lot of things I took from this list, but this is for everyone to discuss. My personal favorites were:

    -Heracross has a lot of awesome resists
    -Tyranitar is the only rock pokemon in OU
    -Flying type moves are rape
    -Fire type moves will make more pokemon run for the hills than any other type
    -Ice/Fight is pretty much the best combo for OU, it hits half of OU for SE damage with no overlaps. Only Bronzong, Tentacruel, and Starmie, completely resist it.

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