Analysis of Base Stats

Speaking of this, I'd need a bit of help for this. I'll basically need a list of every Pokemon's movepool listed in a certain way. I want the movepool to be listed as something like this:

Bulbasaur
Poison 90 S
Grass 80 S
Grass 80 P

etc. Those correspond to Sludge Bomb, Energy Ball and Seed Bomb respectively.

I'm thinking about what to do with variable-power moves. Right now, I'd just need moves that deal a constant amount of damage. Oh, and if a Pokemon learns both Tackle and Return, say, you don't need to list Tackle of course. Also, don't mind accuracy.

EDIT: I need this information only for the fully evolved Pokemon, not for all of them.
Hey, I could do this, but what about hyper beam-esque moves and other 2-turn moves (Solar Beam, Skull Bash etc)? And what about stuff like Leaf Storm?
 
Time Mage: What exactly determines which stat will be used the most? Isn't that the movepool? Look at Flareon. It has 130 Attack stat, yet the strongest move it gets to use it is Return (which does more than Fire Fang). Since this doesn't consider the movepool at all, the best thing I could do is make them do the same amount of damage using both stats. Now, when the movepool is considered, I'll take that into consideration of course.
Most of the time, the chosen stat is the highest one, and that's specially true when there is a big imbalance among the two. I know it is not always true, but since you're going to take movepools (and I'd say, even better, specifically the moves considered in the analysis) into account in the next step, if you want the ranking in your first post to have any significance at all, it should take into account every purely stat-related information.

If this calculations are simply a middle step in a more complex one, and have no significance by themselves, then I really don't see the point of the ranking you posted.

Anyway, this disagreement seems to be too fundamental to be resolved, so I'll leave it here. I respect your work here a lot, and I'm only aiming to make it even better, but some times one has to know when a criticism won't be accepted, and stop there. And I'm not saying you are just ignoring this, of course, but the differences of opinion seem to bee too deep to come to an agreement.

Speaking of this, I'd need a bit of help for this. I'll basically need a list of every Pokemon's movepool listed in a certain way. I want the movepool to be listed as something like this:

Bulbasaur
Poison 90 S
Grass 80 S
Grass 80 P

etc. Those correspond to Sludge Bomb, Energy Ball and Seed Bomb respectively.

I'm thinking about what to do with variable-power moves. Right now, I'd just need moves that deal a constant amount of damage. Oh, and if a Pokemon learns both Tackle and Return, say, you don't need to list Tackle of course. Also, don't mind accuracy.

EDIT: I need this information only for the fully evolved Pokemon, not for all of them.
Wouldn't it be better to incorporate only the moves listed in the analysis? They're there for a reason, that is, those are the most effective ones. Not only the ones in the main movesets, though, but also the alternatives considered.

Anyway, I guess you don't mind accuracy because you're taking it into account, along with critical hit chance, burn/freeze/poison chances, etc. Not necessarily in the next step, of course, but I'm sure you have those in mind.
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
The reason I don't care about variable-base-power moves and moves like Draco Meteor is that I'll probably be going to ban them anyway in my balancing thing.

On second thoughts, please list me the accuracy of the move as well. I was thinking to make all moves 100% accurate, but now I changed my mind. I might change my mind again later, but the accuracy can be listed for now... I'll see if it's useful or not later.

Remember that my ultimate objective is to change the Pokemon's typing/base stats/movepool/abilities so that their rating is much closer to each other. There would still be Pokemon that are better than others, of course, but they won't be better by the same margin as they are right now.

About why I posted this rating system for base stats in the first place, there are two reasons:

1) I think this shows you at a glance which Pokemon have good or bad base stats.
2) I wanted to tell you where I'm headed in my 'balanced Pokemon' project.
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
I'll just need to tell you that if I add Syclant, most base stats' positions will be affected, since you need to remember that a Pokemon's base stat position depends on all the other Pokemon's base stats. If you insert a new Pokemon, the other Pokemon's positions are affected. I'll do a separate chart with Syclant and the new Ghost/Fighting Pokemon stats if you want.

I edited the original post, fixing Cresselia's Speed from 75 to 85. This had the effect of it jumping 5 positions up to #18 from #23, since she scored 3 more with an improved speed (meaning it beat 3 more Pokemon). Other Pokemon were also affected, namely:

Mew, Jirachi, Manaphy, Shaymin, and Celebi lost one place.
Suicune, Heatran and Dragonite lost one place.
Latias, Moltres and Deoxys gained one place.

Charizard's HP fix from 79 to 78 didn't affect any Pokemon's position.

Thanks to sandman for pinpointing these mistakes.
 
70/ 116/ 70/ 114/ 64/ 121 is SYCLANT.

I'll get MUMRA's stats now.

EDIT: Voila: 90 HP / 105 Attack / 90 Defense / 65 Sp. Attack / 110 Sp. Defense / 65 Speed
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
Here's the Top 100 Pokemon containing Syclant and Pokemon#2. Look how the other Pokemon's positions changed because of the introduction of two new Pokemon:

Code:
Rank    Name      HP  Atk  Def  SpA  SpD  Spe  BST  Rating
-----------------------------------------------------------
  1    Arceus    120  120  120  120  120  120  720    499
  2    Mewtwo    106  110   90  154   90  130  680    497
  3     Lugia    106   90  130   90  154  110  680    496
  4    Slaking   150  160  100   95   65  100  670    493
  4   Regigigas  110  160  110   80  110  100  670    493
  6    Palkia     90  120  100  150  120  100  680    492
  7   Giratina   150  100  120  100  120   90  680    490
  8     Ho-oh    106  130   90  110  154   90  680    488
  8    Dialga    100  120  120  150  100   90  680    488
 10    Kyogre    100  100   90  150  140   90  670    487
 10   Rayquaza   105  150   90  150   90   95  680    487
 12    Groudon   100  150  140  100   90   90  670    484
 12   Garchomp   108  130   95   80   85  102  600    484
 14    Latios     80   90   80  130  110  110  600    481
 15   Tyranitar  100  134  110   95  100   61  600    478
 15   Salamence   95  135   80  110   80  100  600    478
 15    Darkrai    70   90   90  135   90  125  600    478
 18   Cresselia  120   70  120   75  130   85  600    477
 19      Mew     100  100  100  100  100  100  600    476
 19    Celebi    100  100  100  100  100  100  600    476
 19    Jirachi   100  100  100  100  100  100  600    476
 19    Manaphy   100  100  100  100  100  100  600    476
 19    Shaymin   100  100  100  100  100  100  600    476
 24    Suicune   100   75  115   90  115   85  580    473
 24    Latias     80   80   90  110  130  110  600    473
 26    Raikou     90   85   75  115  100  115  580    471
 27    Zapdos     90   90   85  125   90  100  580    470
 28     Entei    115  115   85   90   75  100  580    469
 29    Moltres    90  100   90  125   85   90  580    468
 29    Heatran    91   90  106  130  106   77  600    468
 31   Dragonite   91  134   95  100  100   80  600    467
 32   Metagross   80  135  130   95   90   70  600    464
 33     Uxie      75   75  130   75  130   95  580    463 (+1)
 34   Articuno    90   85  100   95  125   85  580    461
 34    Deoxys     50  150   50  150   50  150  600    461
 36    Snorlax   160  110   65   65  110   30  540    458
 37   Registeel   80   75  150   75  150   50  580    457
 38   Arcanine    90  110   80  100   80   95  555    456
 39    Mesprit    80  105  105  105  105   80  580    455 (-1)
 39    Azelf      75  125   70  125   70  115  580    455 (+1)
 41   Deoxys-D    50   70  160   70  160   90  600    454 (+1)
 42    Lapras    130   85   80   85   95   60  535    445
 43    Walrein   110   80   90   95   90   65  530    442
 44    Kingdra    75   95   95   95   95   85  540    440 (-1)
 44    Milotic    95   60   79  100  125   81  540    440
 44   Regirock    80  100  200   50  100   50  580    440 (-1)
 47   Charizard   78   84   78  109   85  100  534    439 (+2)
 47  Typhlosion   78   84   78  109   85  100  534    439 (+2)
 47   Porygon-Z   85   80   70  135   75   90  535    439 (+2)
 50   Swampert   100  110   90   85   90   60  535    438 (-2)
 50    Regice     80   50  100  100  200   50  580    438
 50   Magmortar   75   95   67  125   95   83  540    438
 53   Gyarados    95  125   79   60  100   81  540    437
 54  Electivire   75  123   67   95   85   95  540    436 (+1)
 55   Infernape   76  104   71  104   71  108  534    435
 56   Deoxys-S    50   95   90   95   90  180  600    433
 57   Togekiss    85   50   95  120  115   80  545    431
 58   Torterra    95  109  105   75   85   56  525    430
 59   Blaziken    80  120   70  110   70   80  530    429 (+1)
 60   Mamoswine  110  130   80   70   60   80  530    428 (-1)
 61    Crobat     85   90   80   70   80  130  535    427
 61   Hippowdon  108  112  118   68   72   47  525    427
 61  Lickilicky  110   85   95   80   95   50  515    427 (+1)
[B] 64    Syclant    70  116   70  114   64  121  555    426 (NEW)[/B]
 65   Sceptile    70   85   65  105   85  120  530    424 (+1)
 66    Flygon     80  100   80   80   80  100  520    422 (+1)
 66   Rhyperior  115  140  130   55   55   40  535    422
 68  Feraligatr   85  105  100   79   83   78  530    421 (+1)
[B] 68   Pokemon#2   90  105   90   65  110   65  525    421 (NEW)[/B]
 70   Empoleon    84   86   88  111  101   60  530    419 (+2)
 71   Blastoise   79   83  100   85  105   78  530    418 (+2)
 71    Luxray     80  120   79   95   79   70  523    418 (+2)
 73   Vaporeon   130   65   60  110   95   65  525    417 (+2)
 73   Magnezone   70   70  115  130   90   60  535    417 (+2)
 75    Umbreon    95   65  110   60  130   65  525    415 (+2)
 75    Yanmega    86   76   86  116   56   95  515    415 (+2)
 77    Starmie    60   75   85  100   85  115  520    414 (+4)
 78    Lucario    70  110   70  115   70   90  525    412 (+2)
 79    Espeon     65   65   60  130   95  110  525    410 (+2)
 80   Meganium    80   82  100   83  100   80  525    409 (+2)
 81   Venusaur    80   82   83  100  100   80  525    407 (+2)
 82   Exeggutor   95   95   85  125   65   55  520    406 (+2)
 83   Porygon2    85   80   90  105   95   60  515    404 (+2)
 83    Glaceon    65   60  110  130   95   65  525    404 (+2)
 85    Leafeon    65  110  130   60   65   95  525    401 (+2)
 86    Golduck    80   82   78   95   80   85  500    400 (+2)
 86    Gallade    68  125   65   65  115   80  518    400 (+1)
 88    Flareon    65  130   60   95  110   65  525    398 (+2)
 89  Aerodactyl   80  105   65   60   75  130  515    396 (+2)
 90   Nidoking    81   92   77   85   75   85  495    395 (+2)
 90    Pinsir     65  125  100   55   70   85  500    395 (+1)
 92   Gardevoir   68   65   65  125  115   80  518    394 (+3)
 92   Tangrowth  100  100  125  110   50   50  535    394 (+1)
 94    Machamp    90  130   80   65   85   55  505    392 (+1)
 94    Kingler    55  130  115   50   90   75  515    392 (+1)
 94    Scyther    70  110   80   55   80  105  500    392 (+2)
 97   Rapidash    65  100   70   80   80  105  500    391 (+4)
 97    Slowbro    95   75  110  100   80   30  490    391
 97    Jolteon    65   65   60  110   95  130  525    391
 97    Scizor     70  130  100   55   80   65  500    391 (+4)
Number in brackets is the number of places it fell (positive) or climbed (negative) from the originally posted list. Those having no number in brackets stayed where they were previously.
 
Now that my brain has had some time to grind this through my gears and actually understand it, I have a few things to say.

Before that though, I'm curious: Did you consider things like Huge/Pure Power or Hustle in these "battles?"

My biggest issue with this is that Att and SpA are treated equally in the rating in all cases. This ignores 2 key aspects of attackers:
  1. Pokemon with one really good offensive stat and one really bad one (like Alakazam and Heracross) can choose to focus entirely on the good stat and totally ignore the bad one, becoming a pure attacker and not allowing the bad offensive stat to drag it down.
  2. Mixed attackers can choose to hit an opponent exclusively with either a Physical or Special attack depending on which would do more damage.
To rectify these problems I see, I propose this as a possible solution:

Split each Pokemon into 3 different "classes." Nomenclature and the behavior of the classes would go like this:
  • P-class Pokemon would have P_ preceding the name, such as P_Infernape. This denotes a purely Physical Pokemon, and only the Att stat would be used in the calculations of the simulated "battles."
  • S-class Pokemon would have S_ preceding the name, such as S_Infernape. This denotes a purely Special Pokemon, and only SpA stat would be used in the calculations of the simulated "battles."
  • M-class Pokemon would have M_ preceding the name, such as M_Infernape. This denotes a Mixed attacking Pokemon, and both Att and SpA have a chance to be used in the calculations. In a battle with an M-class Pokemon, the Pokemon would determine whether attacking with the Att or SpA stat would do more damage to the opponent, and would use that stat for the calculation of the simulated "battle."
Under these rules, an M-class Pokemon should always be ranked ahead of its S-class and P-class counterparts. I believe this because M-class Pokemon can choose which stat is best for any given battle, while P- and S-classes are stuck using a potentially disadvantageous stat (such as S_Heracross vs Blissey).

Furthermore, let us consider how a Pokemon's different classes could be ranked in relation to each other.

Consider Weavile, a Pokemon with large physical power and pitiful SpA. A Pokemon like this would gain a comparatively small advantage in the M-class since the SpA might not be large enough to be stronger than Att in many cases, even against SpD weak foes like Hippowdon. This would mean that the M-class and the P-class would be ranked relatively close to each other, with the S-class ranked far below. The opposite would be true for a SpA dominant Pokemon like Alakazam.

For Pokemon whose offesive stats are fairly close to each other, such as Salamence, Lucario, Infernape, etc., there would be a large advantage to using a Mixed moveset, taking advantage of being able to hit opponents strongly on their weaker defensive stat. Therefore, the M-class set should be ranked much higher than the P- and S-class sets, with the S- and P-classes ranked comparatively close.

If that prediction holds true, then by comparing a Pokemon's 3 class' position in relation to each other, you can roughly gauge how much of a general advantage a Pokemon can have by using a Mixed or pure moveset.

Granted, this could only be used to gauge general advantages. There can be specific advantages for a Pokemon to run a slightly mixed moveset, even if the Pokemon's M-class is ranked fairly low in relation to one of the pure classes. Heracross can benefit from using Hidden Power Ice to hit Gliscor, one of its main counters. That is a specific advantage though, because Close Combat would hit most things that aren't Gliscor much harder, even something like Hippowdon.
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
I don't agree with this, and I'll explain why.

The problem here is the movepool. It's useless to have 130 base Attack if your physical movepool is shit (ask Flareon, or Absol in ADV). You're assuming that the higher stat would be used more often than not, but this is not always the case.

Now, as I have always said when such points were raised, when the movepool will be considered, this point will be automatically fixed. Until then, it's easier to just assume that both the attacking sides will be used equally well, and leave it at that.
 
I just registered to say that I'm interested in this sort of analysis, even if it's not strictly relevant to real battles. A couple days ago (before I saw this thread), I calculated (just to satisfy my curiosity) what the average weakness/resistance coefficient was for each type pairing (with about 10 lines of MATLAB code, not by hand :)):

http://img136.imageshack.us/img136/9453/defensematrixos0.jpg
and the same thing for attacking types:
http://img126.imageshack.us/img126/6866/attackvectorfm5.jpg
(a little bit hairier, had to use a 3-dim array with coefficient as a function of attack type and both defending types)

So in a perfectly balanced pokeworld, Ground is the most powerful attacking type and Poison the weakest. Ice/Rock would be the worst defensive typing and Ghost/Steel the best (with Ice/Rock taking over twice as much damage as Ghost/Steel on average). Neither type even exists in the game, and it doesn't account for STAB, move pools and such, but I thought it was interesting. Anyway, I'm proficient in this kind of linear algebra and programming, so if you have any ideas you need help running the numbers on, drop me a PM.
 
The reason I don't care about variable-base-power moves and moves like Draco Meteor is that I'll probably be going to ban them anyway in my balancing thing.

On second thoughts, please list me the accuracy of the move as well. I was thinking to make all moves 100% accurate, but now I changed my mind. I might change my mind again later, but the accuracy can be listed for now... I'll see if it's useful or not later.

Remember that my ultimate objective is to change the Pokemon's typing/base stats/movepool/abilities so that their rating is much closer to each other. There would still be Pokemon that are better than others, of course, but they won't be better by the same margin as they are right now.

About why I posted this rating system for base stats in the first place, there are two reasons:

1) I think this shows you at a glance which Pokemon have good or bad base stats.
2) I wanted to tell you where I'm headed in my 'balanced Pokemon' project.

Perhaps you could rank each move by average damage output over two or three turns; for example, Fire Blast would be:
(120*.85) + (120*.85) for two turns.
Draco Meteor would be:
140*90 + (140/1.5)*90
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
Bump!


I thought Kingler has 50 SpDef?
Well, I made a mistake there it seems.

Theo: You might wish to look at this thread to see my take on attacking type effectiveness.

DarkFlagrance: I'm sure that whoever invented moves like Draco Meteor had your average damage thing in mind, but that's not how it works in battles. If a single Draco Meteor OHKOes a Pokemon, it's OHKOed, even though it will deal half of the damage the next turn. Draco Meteor's power is 140, then 70 and then 46.67. The average damage is 85.56, but if Draco Meteor hit for 85.56, 85.56 and 85.56, it's probable that the first hit doesn't OHKO if you work it out that way, but, as you know, this isn't how it should be calculated. Since this will be a bit difficult to calculate, I'd rather not even consider such moves.
 
So X-Act, what are you trying to show from this list? You did a great job with your mathmatical formulas (I actually have no idea what all those formulas meant, so I'm making the safe assumption that you did a great job), but does this list show us anything different than just adding up every pokemons BST and then listing them in order? It certainly has many of the same errors that listing pokes in order of BSTs would have, like listing Blissey and Deoxys-A and skarmory at the bottom of the list and putting shaymin, entei and registeel at the top of of the list.

If you plan on making a continuation of this list, I would like to suggest excluding a pokemons useless stats. For instance, if you ignore Blissey's useless speed and attack stats, it would definitely rate a lot higher on the list. Don't know how you would do that for some pokemon and not others though.
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
It certainly has many of the same errors that listing pokes in order of BSTs would have, like listing Blissey and Deoxys-A and skarmory at the bottom of the list and putting shaymin, entei and registeel at the top of of the list.
How are those errors? That place is where they belong. With a post such as that, you obviously did not understand what the chart is saying.

You mentioned that 'putting Shaymin at the top of the list is an error'. But you presumably did not complain about Manaphy, Celebi, Jirachi and Mew, who have the exact same stats. So, surely if a Pokemon has the same base stats as another one, their base stats are equally effective? How is Shaymin an error and Mew not an error?

You're putting prejudices on a Pokemon. Entei sucks because its typing sucks and it has an awful movepool. Create a Dark/Ghost Pokemon with Entei's stats and a great movepool and it would be uber, I assure you. Entei's stats are very good: it certainly is not stats that are Entei's problem, so my system listed Entei's stats where they belong: among the best.

Deoxys-A is uber only because its movepool allows it to OHKO Pokemon more often than it is OHKOed. But my system gives the Pokemon the same limited movepool... so down the list it goes. My system also gives Blissey 56 Defense stat to be fair with every Pokemon. With such a pathetic Defense stat, it can't survive anything with a good Attack stat, so down the list it goes as well.

I can continue ranting on and on, but I'll conclude by repeating what I probably repeated in practically every post in this thread: THIS CHART IS ONLY LOOKING AT THE BASE STATS OF A POKEMON: NOTHING ELSE.
 

X-Act

np: Biffy Clyro - Shock Shock
is a Site Content Manager Alumnusis a Programmer Alumnusis a Smogon Discord Contributor Alumnusis a Top Researcher Alumnusis a Top CAP Contributor Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnusis an Administrator Alumnus
Fixed Kingler's SpD from 90 to 50. This caused it to drop by a massive 91 places in the ranks, going from #94 to #185.
 

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top