ELO System

[PedroLindoUnico]

Hello everyone, I come here to express my opinion about this new Elo system.

 

In my opinion, this system is good but it is not effective in games like pokemon which is extremely based on luck.

Let's take a few examples, imagine if I'm fighting and with each victory I'm earning 3 points, but after winning 2 in a row I lose one because of a critical hit or miss causing me to lose 7 points, in other words, 2 difficult victories are irrelevant because of a ridiculous match.

 

Good is that, remembering is just my opinion.

[XelaKebert]

If you go into a battle depending on only luck you are going to lose every time. Critical hits and accuracy are only two aspects of the battle. On top of that they are also the least reliable. You can't say an ELO system doesn't work only because of those elements. Crits and accuracy are nice when they work in your favor and it's easy to get salty when they work against you, but you are also missing out on other aspects such as weather conditions, abilities, and status ailments. Then you have stat buffs and reductions to factor in for a player trying to setup for their strategy. The most successful players make the best use of the reliable aspects of the system. The number of wins and losses a player has should not be the sole factor for rankings ever. Any player can rack up wins if they only get paired against opponents weaker than them. The ELO system is a means of weighing skill in the equation based on strength of opponent.

Even Poker has a modified ELO system and it is based much more on luck than Pokemon.

[PedroLindoUnico]

 

Understanding your point, but we all know that in pokemon, luck outweighs the skill factor. 

I think that in theory this system works but in practice I don't think it is that efficient.

What I think is: if I earn 3 points by battling someone of the same level why not lose 3 if I lose.

 

[rangiras]

Can you explain why Pokémon is "extremely based in luck" asides from just saying "we all know it"? 

[Raichuforyou]

ELO in most competitive games you play is weighted on a system to where it predicts how likely you are to beat your opponent. In most cases, if you lose to someone better than you, you won't lose a ton of points, but if you lose to someone worse than you, you will lose quite a few points. And visa-versa for wins as well.

 

There are luck elements present within the game, but any good team should have a fallback plan if they get hit by a crit, miss their move, etc. The top of the leaderboard for any tier is certainly not random.

[XelaKebert]

On 8/25/2020 at 5:15 PM, PedroLindoUnico said:

Understanding your point, but we all know that in pokemon, luck outweighs the skill factor.

So a 6.25% base chance for a critical hit outweighs skill? That's without any buffs like a Scope Lens, using a move that has an increased chance, or using Focus Energy. Status ailments also factor in, with poison being the most reliable. 

 

As I pointed out, ranked Poker uses and ELO rating as well. Players in Poker don't get any modifiers on their luck at all.

https://poker.rheiagames.com/en/psr#:~:text=Poker ELO %2F Poker Skill Rating&text=New players start with a,reach up to 1%2C250 points.

On 8/25/2020 at 5:15 PM, PedroLindoUnico said:

I think that in theory this system works but in practice I don't think it is that efficient.

What I think is: if I earn 3 points by battling someone of the same level why not lose 3 if I lose.

Literally no ELO system works that way. Just because you both have the same ELO, that doesn't take anything else into account. FIDE chess ratings take a development coefficient into account. So even if two players with a 1200 ELO face each other, they could both have different coefficients, which would also change how many points you would gain from a win and how many points you lose following a loss.

 

https://en.wikipedia.org/wiki/Elo_rating_system#Performance_rating

https://ratings.fide.com/calculator_rtd.phtml