What's new

Matchup numbers - are we doing this correctly?

gitblame

Noob
Quick maths!

So, we like to express matchups numbers as 2 integers that summed up would gives us 10, for example 6-4. It's the same as we said: "6-4 is the most likely score after 10 matches between 2 characters". We could also say that in single game one character has 60% chances of win and other 40%.

How does it apply to tournament setting when we play first to 3? If we treat each match in a set as an independent random event we will get results as below:

6-4 matchup means that one character has 67,656% chances of winning a FT3 set
7-3 - 83,692%
8-2 - 94,208 %
9-1 - 99,144 %

So, remember. When you're on a downplaying fever and want to call some matchup 3-7 you're basically saying that your character has less than 17% chances of winning a set.

We can safely assume that Injustice 2 currently does not have any 8-2 matchup. We probably do not have any matchup in a game that is so bad that we can call it 7-3. And many matchups that are described as 6-4 are not really that bad.
Do we need another way to express matchups?
 

SaucyD0ge

Worst european batman
Quick maths!

So, we like to express matchups numbers as 2 integers that summed up would gives us 10, for example 6-4. It's the same as we said: "6-4 is the most likely score after 10 matches between 2 characters". We could also say that in single game one character has 60% chances of win and other 40%.

How does it apply to tournament setting when we play first to 3? If we treat each match in a set as an independent random event we will get results as below:

6-4 matchup means that one character has 67,656% chances of winning a FT3 set
7-3 - 83,692%
8-2 - 94,208 %
9-1 - 99,144 %

So, remember. When you're on a downplaying fever and want to call some matchup 3-7 you're basically saying that your character has less than 17% chances of winning a set.

We can safely assume that Injustice 2 currently does not have any 8-2 matchup. We probably do not have any matchup in a game that is so bad that we can call it 7-3. And many matchups that are described as 6-4 are not really that bad.
Do we need another way to express matchups?
We can use a sum of 20 in order to achieve more precision.
Example:
Matchup T slightly favors character A over B so instead of a hard 5-5 in favor of A; we can technically use 11-9 in in favor of A.
 

gitblame

Noob
We can use a sum of 20 in order to achieve more precision.
Example:
Matchup T slightly favors character A over B so instead of a hard 5-5 in favor of A; we can technically use 11-9 in in favor of A.
I don't think that putting matchup numbers is a good way at all. Now we just use them as labels. They have nothing to do with how real chances are between two characters. By your example, 11-9 seem like fair matchup where one character has only very slight edge over the other, but mathematically it means in FT3 set that one character has almost 60% chances of winning this. That's significant difference. You would think twice before you would play 11-9 matchup in tournament if you knew that :)

To determine matchups by numbers correctly we will have to gather huge amount of data which is kinda impossible because you will have to find players at exact same level on every matchup and make them play hundreds of matches.
 

JBeezYBabY

Mr. Righteous
Honestly think MU numbers are so unnecessary because ppl look too much into them. Its either if the character looses more against a another character or wins more. THATS IT. Plain, cut, and dry.
 

Vengeance135

Saltiest Joker Player
Looking at injustice 1 joker to injustice 2, joker actually handles characters better in this game. With that being said. He still does have many disadvantages in most MU's and does suffer from a couple 7-3 MU'S. The ones to note, Dr Fate and Super-Girl. But hey that's way better then his odds vs zod or mmh in injustice 1. Those MU's were absolutely terrible.

But you do bring up a valid point. I think the system based on 10 is a bit more simplistic in what you're trying to say. Though, if it were me, I would just say rather the character wins or not with no numbers involved.

For example:

Joker vs black adam- Adam wins

That's a hell of a lot more simple imo lol
 

JBeezYBabY

Mr. Righteous
Looking at injustice 1 joker to injustice 2, joker actually handles characters better in this game. With that being said. He still does have many disadvantages in most MU's and does suffer from a couple 7-3 MU'S. The ones to note, Dr Fate and Super-Girl. But hey that's way better then his odds vs zod or mmh in injustice 1. Those MU's were absolutely terrible.

But you do bring up a valid point. I think the system based on 10 is a bit more simplistic in what you're trying to say. Though, if it were me, I would just say rather the character wins or not with no numbers involved.

For example:

Joker vs black adam- Adam wins

That's a hell of a lot more simple imo lol

EXACTLY!
 

EntropicByDesign

It's all so very confusing.
I think it would be interesting to take data from, say, top 32 and up from all tournaments so far and lay out the odds. Keep player names out, just "Can vs black Adam happened this many times and these were the results" then do the same but exclude top 32 results and include only top 16.

Now, this isnt going to provide an accurate snapshot of rare or lower tier characters, but it would flesh out the popular matchups at a tournament level fair well.

Matchups are considered at the highest levels of play, but I'd argue that top 16 at any major tournament is going to be top play. Top 32 is too, depending on the tournament.
 

JBeezYBabY

Mr. Righteous
The problem isn't with the system. Its the players looking to much into mu numbers, using them as an excuse to why they are losing or trying to use tournament results as to why a character is good or bad.
SAY IT LOUDER TO THE PEOPLE IN THE BACK!
 
Quick maths!

So, we like to express matchups numbers as 2 integers that summed up would gives us 10, for example 6-4. It's the same as we said: "6-4 is the most likely score after 10 matches between 2 characters". We could also say that in single game one character has 60% chances of win and other 40%.

How does it apply to tournament setting when we play first to 3? If we treat each match in a set as an independent random event we will get results as below:

6-4 matchup means that one character has 67,656% chances of winning a FT3 set
7-3 - 83,692%
8-2 - 94,208 %
9-1 - 99,144 %

So, remember. When you're on a downplaying fever and want to call some matchup 3-7 you're basically saying that your character has less than 17% chances of winning a set.

We can safely assume that Injustice 2 currently does not have any 8-2 matchup. We probably do not have any matchup in a game that is so bad that we can call it 7-3. And many matchups that are described as 6-4 are not really that bad.
Do we need another way to express matchups?
You act like people being informed of this would change anything. Literally everyone still feels the exact same way about their MUs even if they didn't know this. I rarely trust MU numbers because of how inconsistent they are within the community and how inconsistently they are measured. There are some way harder supposed 5-5s than supposed 6-4s in this game and not all supposed 6-4s are the same. People instantly think that if the opponent has different counters to their character that it is instantly a bad mu and several factors like risk and reward, damage, comeback/closing potential, movement etc all often go overlooked. Personally I feel that the system should be taken with a grain of salt and people should try to create a more accurate system than this.
However by the standard of straight which character wins the mu more there are definitely 7-3s in this game and maybe 1 or 2 8-2s.
 
Last edited: