Win rate leaderboard (05.01.2018)

[Imaizumi Kagerou]

Current edition:

https://docs.google.com/spreadsheets/d/1lrSm493tTFml9LBOzuv_WaOIRKE-aEW4BWku401quhw/edit?usp=sharing

Older edition(s):

https://docs.google.com/spreadsheets/d/1r1xHxNWmJVa12L31L5oss0dp01kv5PVMxrrJNfdFbqo/edit?usp=sharing

https://docs.google.com/spreadsheets/d/18s8xdVfXR3FTSNPRQfXjWyY4opwdN1HyJwfyXMyewoI/edit?usp=sharing

Original post:

100% Orange Juice is quite a casual game, most players see it as a way to have fun with friends on weekends by rolling some dices and laughing at each other's misfortunes. Even most experienced players don't take the game very seriously and put their time into collecting cosmetics. "Why try your best to win if the game does not encourage it and a single bad roll can screw you over anyway?" — they think.

However, there are a few of us who still try to make the most sense out of a clown fiesta this game is. Speaking for myself, 100% OJ does awaken my inner statistics nerd and optimiser like nothing else. Some of you might remember the Final Battle matchup table[docs.google.com] (now outdated) i made last year. This time around i'm coming at you with the online win rate leaderboard. It features players which are in top 200 in terms of both online wins and online games played. I'll probably be updating it once in a while:

https://docs.google.com/spreadsheets/d/1r1xHxNWmJVa12L31L5oss0dp01kv5PVMxrrJNfdFbqo/edit?usp=sharing

So, if the game was 100% RNG Juice and didn't depend on user input like a lot of people would jokingly (or even seriously) tell you, everyone would be winning about 25% of their games. It turns out that there is quite some space for tryharding, good decision making and outright Sora spamming in this game. We have a few people sitting at above 40% WR, including our frequent poster Shade, and even one person, Sea Gnash, scratching that 45% mark with exactly 666 wins. About half of the list have their WRs above 30% and only ~10% of it do not get their fair quarter of wins, most likely for the reason of not caring about that at all. And i'm not saying that you must care, mind you. This game can be played in different ways: i like it when things make some sense and most players prefer playing with Home Roulette on top of Random Warp. To each their own.

[Cartography Dee]

I didn't expect to be number 30, since I very often join in progress games by replacing a cpu in a terrible position. I guess it goes to show that decisions do matter to an extent.

[Ultrablockstar]

Very interesting calculations indeed. Also this got me to calculate my own win rate which is 26.66%, with 80 out of 300 games won. (very minimal numbers compared to the people in that leader board XD)

[Sjamsjam]

169 baby woo

that's the s e x number but with a 1 so it's better

[Axel]

I don't understand much for what the Win rate?.

[MiYukki]

Ha, I'm suprised im #35 lol

[risbolla]

if you ask me win rates above 25% is more telling about who you are playing against than how well you are playing. without an actual matchmaking system it's not a meaninful number since you could just stomp pubs or play 2 players vs 2 CPU or something like that. or even just leave any match you're losing in the late game. i don't think any of this is behavior that needs to be encouraged.

[Imaizumi Kagerou]

Originally posted by risbolla:

if you ask me win rates above 25% is more telling about who you are playing against than how well you are playing. without an actual matchmaking system it's not a meaninful number since you could just stomp pubs or play 2 players vs 2 CPU or something like that. or even just leave any match you're losing in the late game. i don't think any of this is behavior that needs to be encouraged.

Fair enough, you can rig your stats like that. I'm going with the assumption that people with enough (over 1000) games tracked wouldn't bother doing something like that for extended periods of time. "Pub stomping" doesn't go too well because low level people tend to not play with high level players. And farming CPUs is just boring. Ragequitting denies you stars (can we have those matter again?) and oranges/candies.

[UnLuckyCat]

Originally posted by risbolla:

if you ask me win rates above 25% is more telling about who you are playing against than how well you are playing. without an actual matchmaking system it's not a meaninful number since you could just stomp pubs or play 2 players vs 2 CPU or something like that. or even just leave any match you're losing in the late game. i don't think any of this is behavior that needs to be encouraged.

Mhm, just like that one guy with 0 wins out of a ton of games played, it can be rigged. Also if you stick with a group of friends and all play at a similar level, you'll each have ~25% win rate no matter how seriously everyone takes it. If all four of you fool around the whole time, you'll each randomly happen to win about the same number of times, but if you all go all in with the pro strats, then you'll each spectacularly outplay each other about the same number of times. Both ways give you all ~25%.

Also take note that there's 4 players per game (including CPUs), so a 50% win rate isn't twice as much as a 25% win rate. If you get 1/2 of the wins, the remaining 3 players share the other 1/2, or 1/6 each. That's actually three times as many games won. Winning twice as often as everyone else in the lobby is actually 2/5, since the others get 1/5 each, for 40% win rate vs 20%x3.

[Iridium]

aw, i was about to ask where i was but it looks like i fell out of the top 200 played after challenges killed my interest for a while
pity since i'd be #27 with 407/1159, and i'm still well within the top 200 wins as a result, to the point where i'd still be at 33% if i lost the ~70 games it would require for me to chart

[Axel]

Is it a good or bad the win rate?

[Iridium]

Originally posted by Pisyun:

Is it a good or bad the win rate?

it's not really that meaningful honestly
in theory it would represent how good you are as a player, but in practice it could also just represent how often you play against noobs with two cpus

[risbolla]

Originally posted by UnLuckyCat:

Originally posted by risbolla:

if you ask me win rates above 25% is more telling about who you are playing against than how well you are playing. without an actual matchmaking system it's not a meaninful number since you could just stomp pubs or play 2 players vs 2 CPU or something like that. or even just leave any match you're losing in the late game. i don't think any of this is behavior that needs to be encouraged.

Mhm, just like that one guy with 0 wins out of a ton of games played, it can be rigged. Also if you stick with a group of friends and all play at a similar level, you'll each have ~25% win rate no matter how seriously everyone takes it. If all four of you fool around the whole time, you'll each randomly happen to win about the same number of times, but if you all go all in with the pro strats, then you'll each spectacularly outplay each other about the same number of times. Both ways give you all ~25%.

yeah exactly. in competitive games you would be measured by a separate rating or placement in a league where winning more or less than your fair share would only be an indication that you are playing out of your league. a well functioning matchmaking system would eventually place you in a win/loss equilibrium where you would win half the time in a game with two equal sides or 1 in 4 matches for games with 4 equal sides like oj. you could then determine your skill from the league you're in, in other words who you're playing against on equal terms

[Cartography Dee]

Imagining OJ with matchmaking and a ladder is pretty funny though. Esports here we come

[risbolla]

im ready for the big leagues

[Soda Sip]

Nice, actually wanted to see winrate leaderboard for quite some time. Never thought I'd be so high with all those losing streaks.