Reparametrised base stats

[magic9mushroom]

For fun, I decided to make a transform for the base stats of RBY Pokemon to make them more intuitive and make Base Stat Totals more meaningful.

(This is basically an attempt to put this and this on the same scale, so that they can be compared. Appropriate props to Icy for the idea.)

The transform I used is

Physical Power = 1344/3673 * (2 * baseAttack + 98) * (1 + 5*baseSpeed/2688) - 107712/3673
Physical Bulk = 1/1194 * (2 * baseHP + 203) * (2 * baseDefense + 98) - 347/597
Special Power = 1344/3673 * (2 * baseSpecial + 98) * (1 + 5*baseSpeed/2688) - 107712/3673
Special Bulk = 1/1194 * (2 * baseHP + 203) * (2 * baseSpecial + 98) - 347/597
Speed = baseSpeed

The constants are there to keep Ditto and Mew at 48s and 100s across the board, respectively. Only valid for level 100.

Anyway, here's a table of the renormalised base stats of the fully-evolved RBY mons, rounded to whole numbers.

Pokemon	P. Power	P. Bulk	S. Power	S. Bulk	Speed	Base Stat Total
Venusaur	81	80	96	90	80	427
Charizard	86	76	87	80	100	429
Blastoise	81	90	83	80	78	412
Butterfree	48	53	77	69	70	317
Beedrill	78	49	49	51	75	302
Pidgeot	81	76	73	73	91	394
Raticate	83	57	56	51	97	344
Fearow	91	63	66	61	100	381
Arbok	83	63	67	61	80	354
Raichu	91	56	91	75	100	413
Sandslash	93	93	56	61	65	368
Nidoqueen	80	87	74	79	76	396
Nidoking	90	76	76	75	85	402
Clefable	67	80	80	88	60	375
Ninetales	79	72	100	87	100	438
Wigglytuff	65	75	49	80	45	314
Golbat	81	70	77	73	90	391
Vileplume	74	79	90	88	50	381
Parasect	82	69	70	69	30	320
Venomoth	68	62	89	79	90	388
Dugtrio	86	45	77	54	120	382
Persian	76	60	72	63	115	386
Golduck	82	77	80	78	85	402
Primeape	103	60	65	60	95	383
Arcanine	108	82	82	82	95	449
Poliwrath	82	92	69	76	70	389
Alakazam	59	49	135	96	120	459
Machamp	115	82	63	73	55	388
Victreebel	98	69	94	90	70	421
Tentacruel	74	69	117	102	100	462
Golem	97	108	53	63	45	366
Rapidash	101	66	84	71	105	427
Slowbro	66	104	70	84	30	354
Magneton	61	73	110	85	70	399
Farfetch'd	63	53	58	54	60	288
Dodrio	109	64	65	58	100	396
Dewgong	69	82	90	92	70	403
Muk	94	85	62	78	50	369
Cloyster	90	116	82	67	70	425
Gengar	71	58	128	96	110	463
Onix	48	95	36	36	70	285
Hypno	71	74	106	102	67	420
Kingler	120	85	53	51	75	384
Electrode	62	64	90	69	140	425
Exeggutor	87	88	111	114	55	455
Marowak	73	85	49	53	45	305
Hitmonlee	114	51	42	42	87	336
Hitmonchan	99	64	41	42	76	322
Lickitung	51	79	55	69	30	284
Weezing	84	94	80	74	60	392
Rhydon	111	116	45	64	40	376
Chansey	14	63	94	181	50	402
Tangela	55	91	92	83	60	381
Kangaskhan	94	89	47	61	90	381
Seadra	67	75	93	75	85	395
Seaking	87	69	77	78	68	379
Starmie	81	72	103	80	115	451
Mr. Mime	51	53	98	70	90	362
Scyther	110	74	62	59	105	410
Jynx	56	46	95	80	95	372
Electabuzz	86	59	88	74	105	412
Magmar	94	59	86	74	93	406
Pinsir	118	83	59	57	85	402
Tauros	102	85	76	70	110	443
Gyarados	117	84	96	98	81	476
Lapras	80	99	88	111	60	438
Ditto	48	48	48	48	48	240
Vaporeon	64	84	101	123	65	437
Jolteon	74	60	115	88	130	467
Flareon	118	60	101	88	65	432
Porygon	56	66	68	69	40	299
Omastar	59	99	103	94	55	410
Kabutops	109	83	71	64	80	407
Aerodactyl	111	69	70	66	130	446
Snorlax	94	99	59	99	30	381
Articuno	84	95	118	111	85	493
Zapdos	91	85	122	111	100	509
Moltres	98	89	119	111	90	507
Dragonite	125	92	96	96	80	489
Mewtwo	115	96	155	141	130	637
Mew	100	100	100	100	100	500

[magic9mushroom]

Notes:

Cloyster's and Onix's BSTs drop through the floor because a ginormous Defense stat + shitty HP doesn't give you as much "real" physical bulk as just adding the HP and Defense stats together.

Mewtwo's Uberness is now even more clearly apparent, while it's equally apparent that Mew's is due to its movepool rather than its stats alone.

Zam/Gengar/Jolteon - the triplet of speedy specialists - have their BSTs shoot through the roof.

This chart seems to weight Speed a bit more than it should in terms of viability. Not sure if I'll screw around with the scaling on that. Halving the "Speed" contribution to BST would be the quick fix. Obviously its contributions to "Power" stats have to be left alone to preserve the value of the list, though. You could also screw around with weighting the non-dominant attack stat 1/2 as well.

[marcoasd]

Nice, but imo you shouldn't add the speed stat to the total, as you take into account the CH rate adding it to the power. The speed still counts, but only to outspeed the opponent. Maybe list them in decreasing speed order in that case.

[magic9mushroom]

Jun 10, 2014 0:47:47 GMT -8 marcoasd said:

Nice, but imo you shouldn't add the speed stat to the total, as you take into account the CH rate adding it to the power. The speed still counts, but only to outspeed the opponent. Maybe list them in decreasing speed order in that case.

Like I said, it should probably be weighted less than 1.

[piexplode]

I'm thinking there might be a way to weight movepools as well, involving basically a sum total of values given for each move, each first given checks for seeing if they're outclassed (so if a mon had Rock Throw and Rock Slide, Rock Throw scores zero), with a multiplier involving STAB, and also the physical or special attack power you already have, and then fiddling with it till it looks sensible, then scaling it to have mew have 100 (and it probably has the best movepool anyway p much so everything else would have less) but for the case of if a pokemon has everything it effectively needs and not much else it could want (e.g tauros) then fiddling it so that generallyit is high enough.. I'm not sure how I'd weigh it as a stat or what I'd even call it but what do u guys think of the concept I'm trying and not really succeeding to propose here. Is it overly complicated and unnessecary or is there some sense in it?

Oh btw really nice job m9m ur work always impresses me (and it's inspired me such that when I have the options to look at theoretical statistics/probability etc. I'll take it)

[cheese]

Haha, nice trolling. There's no way in hell you could assign a statistically significant number to a pokemon's movepool.

Movepool is highly subjective, and the value of a move can vary significantly depending on other factors. For example Counter's ability depends on what pokemon are likely to switch in.

Also how to factor what moves are redundant? Ice Beam isn't redundant if the pokemon has Blizzard, but you wouldn't (usually) run both on one set, meaning that having both isn't a huge boost over having just one of them.

What about the value of the move type? For example Mega Drain is a pretty piss-poor attack, but it's occasionally used because some pokemon are 4X weak to it.

If you want to assign a score to the strength of a pokemon's movepool that's fine, but trying to "discover" the score by assigning scores to each move is ridiculous.

[magic9mushroom]

Yeah, as far as damaging moves go you'd have to do it on a wall basis. E.G. "Does it get Earthquake, STAB Drill Peck, or STAB Psychic? If yes, add 1 to movepool score, as it is capable of effectively damaging Gengar."

[ddx2]

What about assigning points for each maneuver from a set they can learn? As in, the more it learns from this table, the better. Plus bonus points for amount of type/status variability.

[Isa]

I'd say that it doesn't matter at all if my Chansey could learn Self-Destruct, Body Slam or Hyper Beam or not.

[CrapAtRBY]

This is probably more useful as something that indicates how much of a pokemon's viability is due to it's movepool by comparing the ranking to the community rankings (I know the community rankings are not a proper or objective measure) anyway.

Edit:

Google Drive spreadsheet version I made so I could sort the list Hope Magic doesn't mind. I was going to do macros for sorting it but it turns out you need an add on for it in google drive and I imagine everyone here can use basic spreadsheet functions so I've just frozen the first row and column 

[magic9mushroom]

Jun 13, 2014 2:45:43 GMT -8 Isa said:

I'd say that it doesn't matter at all if my Chansey could learn Self-Destruct, Body Slam or Hyper Beam or not.

Well, technically Body Slam has an ultra-tiny niche in being able to para Rhydon and Golem, which Chansey otherwise cannot do. Also if you've got an active low-health Screech Persian and a Chansey vs. lone full-health Rest Jynx, Body Slam on Chansey could come in handy.

[jorgen]

I thought STAB SelfDestruct might have the potential to be handy on Chansey if it got it. But then I ran the calcs. It only does a little over 50% to Alakazam. To Snorlax? Like, 26-30%.

Chansey is such a weenie

[cheese]

Yeah, but think about how powerful her Explosion would be!

[CrapAtRBY]

Is this damage calculator faulty then? Cause I've been using it quite a bit

fusxfaranto.com/rbycalc/main.html

Gives me the following for Chansey on Zam

Selfdestruct | 79.23% - 93.29% (min: 248 max: 292)
Explosion | 98.72% - 116.29% (min: 309 max: 364)

[Isa]

Chansey Selfdestruct vs. Alakazam: 161-190 (51.4 - 60.7%) -- guaranteed 2HKO


I suppose that your damage calc has the increased base power from Gen 2 and onwards. I've never heard of that calc previously but yeah it is flawed, don't use it.
pokemonshowdown.com/damagecalc/ this one is superb though.

[magic9mushroom]

Jun 14, 2014 1:22:04 GMT -8 Isa said:

Chansey Selfdestruct vs. Alakazam: 161-190 (51.4 - 60.7%) -- guaranteed 2HKO


I suppose that your damage calc has the increased base power from Gen 2 and onwards. I've never heard of that calc previously but yeah it is flawed, don't use it.
pokemonshowdown.com/damagecalc/ this one is superb though.

No, the calculator wasn't the problem, crapatrby himself was by manually entering "200" and "250" for Base Power.

[Isa]

puu.sh/9sW9o/b80d46e923.png

No, the calculator IS the problem, crapatrby selected Selfdestruct from the dropdown menu.

[magic9mushroom]

Jun 14, 2014 2:22:41 GMT -8 Isa said:

puu.sh/9sW9o/b80d46e923.png

No, the calculator IS the problem, crapatrby selected Selfdestruct from the dropdown menu.

Oh lol, I didn't see the dropdown menu.

[cheese]

Jun 14, 2014 1:22:04 GMT -8 Isa said:

Chansey Selfdestruct vs. Alakazam: 161-190 (51.4 - 60.7%) -- guaranteed 2HKO

Of course, you can't actually 2HKO with self-destruct because by definition you KO yourself in the process.

[piexplode]

should it be like.. guaranteed NKO then (i.e. never to KO) ignoring those highly frequent chansey crits?

[magic9mushroom]

Jun 14, 2014 3:18:26 GMT -8 cheese said:

Jun 14, 2014 1:22:04 GMT -8 Isa said:

Chansey Selfdestruct vs. Alakazam: 161-190 (51.4 - 60.7%) -- guaranteed 2HKO

Of course, you can't actually 2HKO with self-destruct because by definition you KO yourself in the process.

You can't 2HKO, but you could 3HKO. Delicious bugs.