Saturday, August 26, 2017

Mewtwo, why you shouldn't smoke camel dung when giving advice

I've been reading the same advice from two sources concerning mewtwo as a raid boss.

The main culprits are:


In all likelyhood the idiocy has spread from there as well.


Moronic counters

Rankedboost is the worst by a very wide margin, so to make things abundantly clear -- stay the hell away from murcrow, butterfree and ledian when building your lineup against mewtwo.

Let's go through the basics of how dealing damage works in Pokemon Go. This is an extremely simplified version, and for that reason includes some rounding errors, but it covers the basics.

I'm disregarding:
  • the rounding off mechanism which guarantees any attack to deal at least 1 point of damage. It's very important for primary attacks but less so for charge attacks.
  • the level difference multiplier (level of attacking versus defending pokemon, not level of player)


Assume your attack move does 100 damage against a neutral target:
  • STAB multiplies that damage by 1.2
  • Being super effective multiplies that damage by 1.4
  • Being double super effective multiplies that damage by 2 (1.4 squared)
  • Being not very effective multiplies that damage by 0.7
  • Being doubly not very effective multiplies that damage by 0.5 (0.7 squared)
  • That damage is multiplied by attackers attack stat and then divided by the defenders defence stat

Did you get the last part? Attack divided by defence is a multiplier.

The attack stat is a pokemons base attack stat plus 0 - 15, ie the IV attack value of the pokemon. The same goes for defence.

In the case of mewtwo defence is around 200, but for picking an attacker this isn't all that important. Let's say (most likely incorrectly) that the raid boss version of mewtwo has exactly 200 defence.

Base attack stats for suggested crap:
  • Murcrow -- 85
  • Butterfree -- 45
  • Ledian -- 35

Base attack stats for pokemon which are not very effective against mewtwo (given psychic and fighting attacks):
  • Exeggutor -- 233
  • Alakazam -- 271
  • Machamp -- 234

So, since mewtwo is a psychic type pokemon we're comparing murcrow's supereffective attacks with machamp's not very effective attacks. To give murcrow an even greater advantage we're setting it's attack IV to 15 and machamp to 0.

  • Murcrow 100 attack and machamp 234 attack.
  • Murcrow 1.4 multiplier and machamp 0.7 multiplier


Machamp still does more damage by a more than 15% margin given similar attacks despite being at a 100% disadvantage due to combat move typing. Incidentally machamp's best attacks are better than murcrow's.

We won't even look at the atrocities that are butterfree and ledian.


Moronic moves suggestions

Once again Rankedboost has access to the largest stash of camel dung. They explicitly advice you to go with Stone Edge as the charge move for tyranitar.

Gamepress merely incorrectly states that there's no big difference between Crunch and Stone Edge since Bite is the only attack that counts anyway. So in fact they just lit up the dung at a distance from which they couldn't get a really bad trip.


Let's go through the attacks.

Stone Edge:
  • 100 power
  • Requires 100 energy
  • 2.3 seconds cast time
  • Deals neutral damage against mewtwo (multiplier 1.0)

Crunch:
  • 70 power
  • Requires 33 energy
  • 3.2 seconds cast time
  • Deals super effective damage against mewtwo (multiplier 1.4)

Bite:
  • 6 power
  • Adds 4 energy
  • 0.5 seconds cast time
  • Deals super effective damage against mewtwo (multiplier 1.4)

So 25 Bite attacks builds 100 energy. Now you build energy by taking damage as well, but we're just comparing Crunch and Stone Edge here.

50 Bite attacks would allow you to fire off Stone Edge twice.

According to a Reddit thread each Bite attack will inflict 7 damage against mewtwo given a decent enough tyranitar. This implies that 6 (power) time 1.4 (super effective) times attack divided by defence is 7 after all the rounding off to get an integer as the damage value.


Creating the most beneficial scenario for Stone Edge yields the data below:

Crunch will deal 75 damage given a decent tyranitar.

So, 70 (Crunch power) times 1.4 times X is close to 75. This yields an X value of close to 0.8.

Applying that 0.8 against Stone Edge we see how the 100 power would result in it dealing 80 damage.


Now let's count that damage for Stone Edge:
  • 50 Bite does 350 damage
  • 2 Stone Edge does 160 damage
  • The time required is 25 plus 4.6 seconds, so around 30 seconds in total.
  • 510 damage dealt in total

How much damage can we deal with Crunch in 30 seconds?

25 Bite will yield the energy to fire off 3 Crunch. That takes 12.5 plus 9.6 seconds.

Can we squeeze in a fourth Crunch before we've reached 30 seconds?

9 Bite gives 36 energy. That takes 4.5 plus 3.2 seconds, so in total 7.7 seconds.

In total 29.8 seconds, which we round off to 30.

Damage done with Crunch:
  • 34 Bite does 238 damage
  • 4 Crunch does 300 damage
  • 538 damage in total

A solid 5% extra damage with Crunch and you never risk fainting with more than 33 energy. This is an extremely important difference since Bite is a bloody awful primary attack when it comes to building energy. Statistically a Stone Edge tyranitar is likely to go down with enough energy built up to fire off an extra Crunch.


It's possible to squeeze out more damage by having a better tyranitar, but that merely increases Stone Edge damage from 80 to 81 and Crunch from 75 to 80, which widens the gap to 512 versus 558.

Suddenly we have a 46 damage difference, which is closer to 10%, and the benefits from pushing that button every 33 energy still remains.

Notably I haven't taken into account that you'll never, ever produce useless energy with Crunch, which automatically happens if you fail to fire off an 100 energy charge attack before you would receive any more energy. This further increases the gap between Crunch and Stone Edge.

For a gamer a 10% performance boost is just huge at the end game level. Even 5% extra is an important increase.


Closing Words

To further clarify why poor advice is poor. The difference between a Crunch tyranitar and a Stone Edge tyranitar in this fight is larger than the difference between a Stone Edge tyranitar and a DT/Outrage dragonite versus the mewtwo raid boss.

I haven't read aywhere that it doesn't matter if you pick a tyranitar or a dragonite for mewtwo as a raid boss.

1 comment:

  1. Thank you! I've been trying to explain this but people read the crap and believe it. Then they wonder why they need 15 people for a raid.

    ReplyDelete