Ranking the Important Pokémon Trainers: A Deep Dive

Now that my project of Ranking Every Single Important Trainer in Every Main Game of the Pokémon Series Including the Remakes is done I have a very hefty dataset that I can now do some very fun things with and provide y’all with a fun bunch of little informational tidbits and pseudo-trivia. I will be taking my own word as gospel and stating my opinions as facts. It is my website, I can do what I like.

For reference, here is a contents list of all of the ranked lists!

Here’s a quick list of all the lists in this project!

After all of that we have ranked a grand total of 305 trainers, and I’m largely certain I got every important trainer appearance that I could: every rival, gym leader, team bosses and admins, recurring characters, frontier brains, random sidequest characters etc. etc.

Some characters do not appear in these lists. These include the Saffron City Dojo Master, Cynthia in Generation VII, all the non-Unova trainers from the Pokémon World Tournament and probably more. Whoops.

Each of these trainers has been ranked on five matrices, Difficulty, Personality, Fashion, Rosters, and Creativity. They were also attributed up to five bonus points, or given negative bonus points if they displeased me.

These scores were all subjective and arbitrary, and the bonus points even more-so. For example Lance got an extra point in Gen VII because he’s still using Hyper Beam a ton and that pleases me, and Roark loses a point because he seemingly abandons his signature Pokémon for a new ace when you fight him in the Platinum rematch.

At the bottom of this article there will be a link to a Google sheet containing every single individual score and breakdown for bonus points.

So let’s start having a look at some of the interesting pieces of trivia we can glean from having every single trainer all lined up on a single Excel spreadsheet! Firstly, you may want to know who the best and worst trainers are – while you could just scrape through the lists to find that out easily, here’s an even easier way: me telling you!


1. Giovanni from Generation VII

2. Wally from Generation VI

3. Silver from Generation IV

-4. Gladion from Generation VII

-4. Iris from Generation V

6. Leon from Generation VIII

-7. Professor Kukui from Generation VII

-7. N from Generation V

-9. Wallace from Generation III

-9. Agatha from Generation I

-9. Lance from Generation IV


-295. Blaine from Generation III

-295. Falkner from Generation II

-295. Celosia from Generation VI

-295. Tabitha from Generation VI

-299. Bruno from Generation I

-299. Bryony from Generation VI

-299. Shielbert from Generation VIII

-302. Bruno from Generation I

-302. Sordward from Generation VIII

304. Aliana from Generation VI

305. Proton from Generation IV

The Team Flare admins really never stood a chance, did they? And look at poor Proton there. He’s unique in that thanks to his scores of 1 in all five normal categories and his negative bonus point thanks to having less personality than his unnamed Gen II counterpart, he actually has less points than what I’d assumed was my enforced minimum of 5. Way to go, Proton…

We can also have a look at some more sub-categories of trainer to find some interesting tidbits. Let’s have a look briefly at each Pokémon type, seeing how their type-specialists did in this project. For reference, I consider a trainer a type specialist if they are considered one by the games themselves, or if they have teams of two or more Pokémon that all share a type. There are some exceptions where type-specialty is rather implied and I’ve just gone with it, such as Benga being a Dragon-specialist despite owning a Volcarona etc.

What I think will be interesting is comparing the scores within these type categories against the overall averages of the entire cohort – we can compare how type specialists do in the grand scheme of things. So here, for your reference:

  • Average difficulty score: 2.9
  • Average personality score: 3
  • Average fashion score: 3.2
  • Average roster score: 3.5
  • Average creativity score: 3
  • Average amount of bonus points: 1.1
  • Average total score: 16.8

Just in case you were wondering, I did take the liberty of working out the median score and the most average trainer out of the entire lot. The median score of these trainers is 16.5, which is a score held by 14 separate trainers – and it just so happens that the most average trainer lies among this group! To find her, I simply calculated how far each trainer’s individual scores were from the cohort averages that are listed above. After adding all of these averages up, we can see who the most average trainers are and which trainers are furthest from the average.

The biggest outliers are quite obvious to find, they’ll be found at the very top and the very bottom of the list. Proton is the furthest from average, followed by Giovanni, who is closely tailed by Wally and Aliana. Seeing the most average trainers however is far more interesting though – they were all quite middling but not necessarily memorable or unmemorable. The top three most average scoring trainers are Generation I Blaine, Generation IV Saturn and most average of all, Generation III Lorelei. They just… exist, I guess.

Bug icon SwSh.png

No. of trainers: 7

Best trainer: Alder

Worst trainers: Aaron, Bugsy (II)

Average score: 16 (SD = 1.72)

The Bug-type specialists are largely very uniform in their showing, though they do have a quite high average roster score, which I’ll guess is due to Game Freak finally discovering how to build a Bug-team halfway through Gen IV.

GO Dark.png

No. of trainers: 10

Best trainer: Piers

Worst trainer: Sidney (III)

Average score: 16.8 (SD = 3.7)

These trainers definitely benefited a lot from the passage of time: as more Dark-type Pokémon were added into the games, they just became cooler trainers. As you may expect, they have quite a high average roster score, at 3.8

GO Dragon.png

No. of trainers: 16

Best trainer: Iris

Worst trainer: Drasna

Average score: 21 (SD = 2.7)

It was pretty obvious that the Dragon-type would have a great showing what with so many of its trainers being so important, but they impressively score well above average in every single category, and even the worst trainer has an impressive score of 15.5.

Electric icon SwSh.png

No. of trainers: 11

Best trainer: Elesa

Worst trainer: Lt. Surge (III)

Average score: 14.5

You may be thinking that’s a low average score, and you are right. This is because Lt. Surge alone makes up the bottom five trainers of this type, with his Gen III incarnation being the worst and Gen I, II and IV tied at the best, yet none of them getting more than 13 points. Tough show, dude.

GO Fairy.png

No. of trainers: 4

Best trainer: Bede

Worst trainer: Valerie

Average score: 20.3

It’s hard to extrapolate what a possible ‘future of Fairy-type’ might be based on the information gathered in this project as there’s just so few trainers. I count Bede as a Fairy-type specialist as that’s the type he starts with – he doesn’t become a Psychic-specialist until his big promotion.

Fighting icon SwSh.png

No. of trainers: 15

Best trainer: Korrina

Worst trainer: Bruno (I)

Average score: 14.3

Bruno being the worst trainer is… not surprising. Where the type really falters however is in both difficulty and personality – a lot of the trainers are just a bit… boring. Interestingly, the only Fighting-type specialist who isn’t a Gym Leader nor a member of an Elite 4 is Mustard from Gen VIII.

Fire icon SwSh.png

No. of trainers: 15

Best trainer: Malva

Worst trainer: Blaine (III)

Average score: 14.6

While not as bad as Lt. Surge, Blaine impressively makes up both the three bottom placements of this type and is simultaneously the third-best trainer thanks to his Gen I incarnation. While not bad, Fire’s score is brought down by the relative ease with which most of its trainers can be dealt with.

Flying icon SwSh.png

No. of trainers: 6

Best trainer: Kahili

Worst trainer: Falkner (II)

Average score: 16.3

While there are 6 specialists for Flying-type, two of them are repeated so this type is quite comparable to Fairy. It’s actually a bit shocking that a type that has been present since the beginning has so few trainers – it’s not like it’s a boring type.

Ghost icon SwSh.png

No. of trainers: 11

Best trainer: Agatha (I)

Worst trainer: Phoebe (III)

Average score: 18

Ghost trainers are bloody cool. All but three of the trainers have scores well above average, but Phoebe’s scores are just so low as to bring the type down quite a bit, without her the average score would be up by a whole .5 points.

Grass icon SwSh.png

No. of trainers: 10

Best trainer: Milo

Worst trainer: Erika (VII)

Average score: 16.3

So, the Grass-type trainers are fashionable, eh? Their average fashion score is a whopping 3.7 and they get a whole 4 for their fight rosters. I definitely have some favouritism for a nice-looking trainer with a nice-looking team. Plus, Erika definitely bumps up the score thanks to always looking great.

Ground icon SwSh.png

No. of trainers: 7

Best trainer: Giovanni (VII)

Worst trainer: Hapu

Average score: 18.9

With Giovanni making up more than half of the trainers in this group thanks to his appearances in Gens I, III, IV and VII it is no surprise that the average score is quite high. Notably, the Ground-type has the highest average amount of bonus points, so uh, go them?

Ice icon SwSh.png

No. of trainers: 13

Best trainer: Lorelei (I)

Worst trainer: Pryce (II)

Average score: 15.1

You know, it says a lot about how abysmal Pryce was in Gen II that he’s got a lower score than Sina – a trainer whose team you could breathe upon and cause a few casualties.

Normal icon SwSh.png

No. of trainers: 7

Best trainer: Norman (III)

Worst trainer: Ilima

Average score: 20.1

Very strong scores across the board for the Normal-type, it seems. They may be Normal but they are far from average, enjoying by far the highest average difficulty at a mean 4.1/5. They also have the biggest amount of difficulty walls, with Norman and Whitney, and to a lesser extent Lenora and Cheren. Impressive.

Poison icon SwSh.png

No. of trainers: 13

Best trainer: Koga (II)

Worst trainer: Proton

Average score: 14

Poor Proton. He’s the butt of this entire project huh? What’s neat is that Plumeria is only half a point away from tying with Koga, and it would have been cool to see a team admin in both slots for a type. As a type, Poison suffers from predictably low average difficulty scores.

Psychic icon SwSh.png

No. of trainers: 15

Best trainer: Lucian

Worst trainer: Sabrina (III)

Average score: 15

I was honestly a bit shocked when I realised that there are more Psychic specialists over the series than types like Grass and Rock. Psychic’s shortcomings appear to be mostly in their personality scores though they also have a real tendency to not accrue any bonus points which is to their overall detriment.

Rock icon SwSh.png

No. of trainers: 11

Best trainer: Olivia

Worst trainer: Roxanne (VI)

Average score: 13

Out of the eleven trainers who specialise in the Rock-type, eight of them are the first Gym Leader of their region. One is the second. Pokémon loves throwing trainers at Rock Pokémon as their first challenge don’t they? These trainers are quite below average in general but they suffer the most in difficulty and creativity – we all know how to take Rocks on by now.

Steel icon SwSh.png

No. of trainers: 11

Best trainer: Steven (VI)

Worst trainers: Byron, Colress (VI)

Average score: 19.5

Steel is one of the higher-scoring types and it’s kind of fitting – the only trainers I’d consider kinda boring are Byron and Wikstrom, all of the others have some serious effort behind their development. Makes sense that they have really high personality and fashion scores as a result.

Water icon SwSh.png

No. of trainers: 17

Best trainer: Wallace (III)

Worst trainer: Misty (I)

Average score: 18.1

There really is just so much water in Pokémon. The most trainers, the most Pokémon, and a ton of the blue stuff in every region. When you look at the trainers though, there’s probably the biggest amount of diversity! You have young girls, older men, camp fashionistas and even a masked wrestler. I love the Water-type.

Well that was interesting, wasn’t it? No? Oh well. Let’s instead have some pretty graphs imported directly from a PivotTable.

As we can see, each generation largely follows the same trends, but there are some standouts. What you’ll notice is that Gen V gets the higher scores in all five categories. I think this really exemplifies how strong the trainers are in that generation both in building enemy teams and designing/writing compelling characters. Of course, we can’t ignore that Gen V has a lot fewer trainers than the other modern generations and thus the overall score is not brought down by the likes of lacklustre admins, a saturation of superfluous trainers or the fickle nature of remake trainers.

The data also supports some hypotheses too, whether held only by me or commonly throughout the fanbase. For example it’s a very common thought that Gen II was an extremely easy generation and this is held up by the difficulty score. Personality peaks with Gen V and Gen VIII, and these are the generations where I think the most effort was put into the actual narrative, whether through plot writing (Gen V) or character writing (Gen VIII). Creativity falters hugely in Gen VI and Gen VII and this is almost certainly due to how many punches were pulled when designing enemies – there are just too many allowances provided to the player.

Of course, as any statistician/psychologist/person who has done a questionnaire will tell you, this is all complete tosh and basically a vanity project. We obviously don’t know if these opinions would hold up when looking at the wider Pokémon community. Perhaps that is a project for another day…

What I can do as a failed scientist is use a statistical package to try and quantify some of my opinions about Pokémon trainers. For example, I can test to see if my rating of a trainer’s creativity significantly correlates with my rating of their difficulty. After running this test, I did find that there was a large correlation between the two scores, r(303) = .52, p < .000! We can illustrate this with a rather boring chart, like so:

Thanks to the nature of the project itself, it does turn out that every single possible correlation between the five main categories turns out to be significant. This is because I am only one person and it’s so obvious that my opinions would end up having very standardised correlations. I think it would be quite interesting to draw from a far larger sample… Hmm… Watch this space.

In the mean time, that’s it. Am done. Am finished. No more Pokémon for me for a while. Carry on.

Link to the Google Sheet containing a breakdown of scores

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s