Statistics for Melee

[Mew2King]

http://clashtournaments.com/?p=637

 

[Zone]

http://clashtournaments.com/?p=637

This is such a great deed. Thank you lol.

 

[PGH Carroll]

What about

Most Dunks per match/set
Most Johns per match/set
Most Showers per weekend tourney
Most Swear words per match/set (commentary)
Most SD because of Randell (I WOULD LEAD THE LEAGUE)

 

[JKJ]

DUUUUDEE
holy F*** Yes

 

[thespymachine]

I... don't think M2K's brilliant info-dump of Melee is necessarily helpful to the cause JKJ has posited, but I sure am thankful for the contribution and the affect it has had on the Smash community.

 

[thespymachine]

Currently unemployed, so I may be able to work on some stuff for this.
I'm gonna work on The Next Episode top 8.

If anyone would like to help, pick any major tournaments from this year and do the top 8 as well.

Here's the necessary info:
SL = Stock Loss (KO or SD)
Type = KO: Combo Ender (Combo), Edge-Guard/Gimp (EG), Edge-Hog (EH), and Raw (which would be any on stage, neutral game KO that isn't a combo, like a KO from a read or random attack)
= SD: Purposeful (Pur), Accidental (Ac), and Circumstantial (Cir) [Like if you are baited into airdodging and fall to your death, or something]
% = Percentage SL occurs
Place = Placement of SL on deathzone: Left, Right, Top, Bottom
Move = Attack that caused SL (only applicable to KOs)

Brkt Spot = Part of Bracket and which Round the game occurs
Prj FH = Projectile First Hit
Phy FH = Physical First Hit

Also be sure to have specific notes for each match.

Link to GoogleDoc Spreadsheet version

 

[itsbme]

I based this off what you made. For fun, I did a classic set. Below are stats from Moast 3 Finals between Ken and Isai.

The red numbers indicate order in which stock lose occurs for some extra info.

Interesting things to note:

Ken was never first to lose a stock.
Ken varied his edge guard, and didn't Fsmash as much as he seems to now either. Still, he used Fsmash to KO 47% of the time.

Isai ended stocks about 63% of the time with Fair. (Big wow, I know......)
63% is also the average that Isai's KO move was part of a combo. Ken's average KO move that was part of a combo was only 13%.

Ken took a total of 15 stocks.
Isai took a total of 16 stocks.

 

[thespymachine]

I based this off what you made. For fun, I did a classic set. Below are stats from Moast 3 Finals between Ken and Isai.

Awesome!
- I've been meaning to make the form less bland, so I like what you did the darker cells
- the extra row with SL/type/etc is a great idea
- And I think adding the order of stock loss is brilliant
- Before I post what I did, may I ask why you went with the ver. 1 format? (was it just easier to record as you went?)
- Also, what is "IndirKO"?
[I'm trying to get as much feedback as possible to help improve this. Sorry if I'm seeming nosy.]

The little I worked on:

[games 2 and 3 I left Fly's SL stats unempty because I forgot, so he had one stock left in game 2 and 2 stocks left in game 3; also, I forgot to put the match time for game 3]

- Firstly, I realized that I was going for a paper-pencil design with ver. 2, which makes it not as efficient for typing (you end up doing a lot more mousing and tabbing/entering than you really should)
- I started to implement Left and Right on Bottom SL's to help determine ever more accurately where a player died - and it doesn't take much more input for the score keeper
- I do think First-Hit is important (not as important as First Stock Loss), but atm it's difficult for me to think of a better way to keep track of it - because, obviously, physical first hits are going to be more important for MOST of the cast (only really in the case of Samus or Peach does the projectile first hit really 'matter'); because of this, you can see that in games 2 and 3 I don't even put a prj fh because the physical one came first anyways

Getting ideas about ver. 3
Thanks for the input. Keep 'em coming!

 

[itsbme]

Awesome!
- Before I post what I did, may I ask why you went with the ver. 1 format? (was it just easier to record as you went?)

Pretty much!

- Also, what is "IndirKO"?

Indirect KO. It means a move lead to someone being KOd but it wasn't the direct cause(meaning move itself didn't send past blast zone). For example, in one match Ken used nair on Isai while off stage, and Isai was able to up b back towards the stage but couldn't reach the stage.

- I started to implement Left and Right on Bottom SL's to help determine ever more accurately where a player died - and it doesn't take much more input for the score keeper

Yeah I thought of that after I made my chart. It looks weird looking back at it, haha. I'm glad you have some more ideas to utilize. I like this idea!

 

[thespymachine]

Indirect KO. It means a move lead to someone being KOd but it wasn't the direct cause(meaning move itself didn't send past blast zone). For example, in one match Ken used nair on Isai while off stage, and Isai was able to up b back towards the stage but couldn't reach the stage.

Would all edge-hogs be considered indirect-KOs then? I see that there were no edge-hogs in your set, it looks like, and it seems that all EH's are indirectKOs.
Gonna implement that as well.

 

[thespymachine]

https://docs.google.com/spreadsheets/d/1qpIoW7_R48K7g18pZa1d828DMDj7VABYA2gJTCsvqGA/edit?usp=sharing

Tell me what you guys think.

 

[byebye]

I was thinking of this over the weekend. good thing that there's already something like this.

Though I am envisioning more of a generic stats format usable on all fighting games:
aside from what you've listed
whiff and block punish percentage
evade and / or block percentage
projectiles fired per round

then a more situational stat like:
what did the player do after he / she knocks down the opponent (maybe too SF or tekken oriented?)
what did the player do after he sent the opponent flying into the air
DI tendencies. e.g. player A like to do DI in between combos by pressing down 40% of the time. etc.
positioning stat(?). how much percentage of the space and time does a player in control (or own the majority) of the stage. e.g. player A controlled / owned 70% of the map, 80% of the time. - does this makes sense?

Just throwing out ideas that may or may not make sense! I'm interested to see what happens to this!

-------------------------------------------------------------

can this stat collection activity be incorporated within project M? Might there be away to automate some of this?

 

[itsbme]

Would all edge-hogs be considered indirect-KOs then? I see that there were no edge-hogs in your set, it looks like, and it seems that all EH's are indirectKOs.
Gonna implement that as well.

Well, off the top of my head Ken edge hogged a few times. I never thought about it though. Good point. I guess a way to look at it is if say, Player A sends Player B off stage, and Player A edge hogs so when Player B up-Bs, they can't ledge grab and they die as a result of the edge hog, so that was the action that caused the KO, not the move that sent them off stage in the first place.

So I guess you could have these categories.

Edge Hog(EH) = If the result of Player A grabbing a ledge results in Player B being unable to grab the ledge and fall to blast zone.

Knock Out(KO) = If Player A move sends Player B flying past a zone, or if said move bounces Player B off a wall/side of stage into blast zone.

Indirect Knock Out(IndirKO) = If Player A move sends Player B off stage, but not past blast zone, but still sends them far away enough where they can't make it back with any attempt.

Maybe Indirect KO is a good idea, but maybe someone else has a better use for it than what i'm using it for.

As for whether or not EH is a KO or IndirKO, I think you could have a poll. I don't think my opinion alone is valid.

 

[thespymachine]

I was thinking of this over the weekend. good thing that there's already something like this.

Though I am envisioning more of a generic stats format usable on all fighting games:
aside from what you've listed
whiff and block punish percentage
evade and / or block percentage
projectiles fired per round

then a more situational stat like:
what did the player do after he / she knocks down the opponent (maybe too SF or tekken oriented?)
what did the player do after he sent the opponent flying into the air
DI tendencies. e.g. player A like to do DI in between combos by pressing down 40% of the time. etc.
positioning stat(?). how much percentage of the space and time does a player in control (or own the majority) of the stage. e.g. player A controlled / owned 70% of the map, 80% of the time. - does this makes sense?

Just throwing out ideas that may or may not make sense! I'm interested to see what happens to this!

-------------------------------------------------------------

can this stat collection activity be incorporated within project M? Might there be away to automate some of this?

In general, a lot of those stats would take a lot of time to take down as you watched. Personally, my goal was to be able to take down everything as the match played, without overloading the stat-keeper.

However, as you mentioned, I do think it would be a FANTASTIC idea to have those kind of stats kept in the game itself.

Well, off the top of my head Ken edge hogged a few times. I never thought about it though. Good point. I guess a way to look at it is if say, Player A sends Player B off stage, and Player A edge hogs so when Player B up-Bs, they can't ledge grab and they die as a result of the edge hog, so that was the action that caused the KO, not the move that sent them off stage in the first place.

So I guess you could have these categories.

Edge Hog(EH) = If the result of Player A grabbing a ledge results in Player B being unable to grab the ledge and fall to blast zone.

Knock Out(KO) = If Player A move sends Player B flying past a zone, or if said move bounces Player B off a wall/side of stage into blast zone.

Indirect Knock Out(IndirKO) = If Player A move sends Player B off stage, but not past blast zone, but still sends them far away enough where they can't make it back with any attempt.

Maybe Indirect KO is a good idea, but maybe someone else has a better use for it than what i'm using it for.

As for whether or not EH is a KO or IndirKO, I think you could have a poll. I don't think my opinion alone is valid.

It does seem to be dependent on the definitions of KO and iKO.

EH could be direct since it is the move that is the ultimate demise of that stock.
It could also be indirect since the opponent is not interacting with the player at all.
With your example definitions, EH would be neither.

I'm for indirect, because EH is just a special label for 'option removal' - other non-direct moves that don't hit your opponent can (seem) to take away options too, but don't have their own category.
Example: With Falco off-stage, Samus can fire a weak missile at him from stage-level. It takes away Falco's side-b option away, and Falco is forced to drop a bit lower then use Up-B (because Falco will die if getting hit), but Falco is too far away to even reach the ledge and ends up losing his stock.
It was an option removal, that didn't hit the opponent, in the same way holding onto the ledge as your opponent is recovering is.

I suppose also, if we were to use EH as its own category, we would only be using EH as the move used for the SL - but then we could just keep using the move then sent the player off.


Anyone else with input?

 

[byebye]

However, as you mentioned, I do think it would be a FANTASTIC idea to have those kind of stats kept in the game itself.

yeah. can we send a hint to the Project M team and ask them to try and "extract" some information?
if they can just do it, then future PM matches can autoly be stat'ed.

 

[thespymachine]

It would be really cool if they were able to upload said data onto their site as well.

 

[thespymachine]

New setup. Colorful and stuff. Found here: https://docs.google.com/spreadsheets/d/1qpIoW7_R48K7g18pZa1d828DMDj7VABYA2gJTCsvqGA/edit?usp=sharing

 

[thespymachine]

Double-post

 

[thespymachine]

So Fizzi just stepped up and changed the game: https://youtu.be/-W9Sczqtik8