SPL X - Commencement Thread

Status
Not open for further replies.
BD is a stupid system. Instead of BD we should have tiebreak rounds.
Note: as a Naga in Snake I got into playoffs by 1 game of BD. As a BIG I missed tiebreaks on 2 games. Both times it was idiotic and neither thing should have happened.
 

Amaranth

Formerly The Idiot Ninja
is a Pre-Contributoris a Past SPL Champion
If I didn't miss that hbeam vs Tin I would be mentioned in this thread too now and you all know it.
yeah well if I didn't get frozen turn 2 vs troller in week 5... everyone has games or sets to whine about

tbh I think Nails's argument is fallacious, records correlating to price does not mean that the records correlate to how people actually play.
in forming your final record, the form of the opponents you face matters (my set vs beds was by far the worst I played all tournament for example), the actual opponents you get to face matters (lucky bastards all of you who didn't have to face Alexander), and luck still matters a ton.
I'm not making this post to whine about my record, I'm making the point that ultimately personal records don't reliably correlate to skill. Go ask the people with the two best RBY records if they think I was the 6th best RBYer throughout this tournament and come back with the results. And again, I'm only making this about myself because I don't want to put anyone else under scrutiny, but my record is not the only one to be inaccurate.
 
Last edited:

Amaranth

Formerly The Idiot Ninja
is a Pre-Contributoris a Past SPL Champion
if you actually understood my post you'd notice that I was actually arguing that nobody's record accurately reflects their skill, not just mine, and used myself as an example because I didn't want to openly call out anyone else. but I see that me talking about myself has become low hanging fruit for poor sods like you to come out. enjoy the likes, I'll be happy to post more and give you more chances to likegrab so you can sleep at night with your smogon validation
 
I was rooting for the wolfpack but my allegiance is in shambles because FriendOfMrGolem120 is probably the best username in gaming history. Let me explain myself:

First of all it leads to the obvious question, who is mr Golem? I dont know much about the guy but when I put thought to it I cant help but image a golem living in a house resembling a cave. He is no savage tho. He may live a frugal life, but in fact Mr Golem seems rich in more sophisticated qualities. Maybe he is an educated gentleman and a person of high esteem. He could be a prominent academic figure. And even if his social status isn't as celebrated as I make it out to be, he cant be any less than than a respected family man, or else he wouldnt be a mr.

Secondly and in accordance to the above, he seems to be held in high regard by his fellow friends. They put their personal identity aside, opting to define or introduce themselves not by personal traits but by their bond of friendship with him.

Finally Mr Golem seems to have at least 120 friends. All loyal af. Im happy for him.
 
Blunder said, the only thing that should count to your record are wins, which myself as an advanced stats fan in sports doubted really hard, so I did a little exercise today. I calculated the Strength of Schedule (known mostly from college sports), the Relative Percentage Index (NBA metric) and the Expected win Percentage (MBL) of all the SPL players using their W-L records reagrdless of tiers.

This is what came out:
166188


Thought you all might be interested. I do not want to say anyone is better or worse than the other, because you are all way better players than I am and who am I to judge. One good/bad season does not mean anything.
I really enjoy this SPL very very much. So thank you all!

Soooo,
Shoutouts to Fear BLUNDER SANTANA Lavos Charmflash Nails Kickasser Sabella bro fist ...
You are top of the bunch no matter which metric you use, and we also learned that it might not ONLY be win total ;)

Also Tamahome Kickasser Bro Kappa ... you had a damn hard job this SPL.

P.S.: As for the people who understand what I tried here, please tell me if and why my calculations are BS. I am not really an expert. Also I am not happy with the exponent 1.83 in EW%, because it is used for the statistic with runs in baseball and not W-L, but I could not find a better one. Do you?

edit: lmk if you want the same thing for POs. This was only regular season!
 
Here is the Spreadsheet for you all. I was simply amazed how you all reacted and that you took the time to give me some propz :).
First of all, though, I copy pasted the data, not the functions etc, so there is nothing to do for anyone in this sheet but me. The original sheet will stay offline, unless someone wants to and will work with me on whatever that might be. Please contact me on Discord or on Smogon if there is something wrong.
You can search for your personal stats using the Dropdown in the “Find your Stats” sheet. Also, there are rankings for the three advanced stats and a sheet with most of the important data all in one.
Second of all: There was an arithmetic mistake in my first post. Some numbers changed a bit from yesterday, sorry!
@ all the team tour managers: I would love to be part of a team in the future. So maybe you might consider contacting me for some Pokemon John Hollinger stuff :D!
For those who want to understand what I did, I’ll try my very best, so here it goes:

W: wins
G: games
P: player
O: opponent
OO: opponent’s opponents
W%: Win Percentage = W(P)/G(P)
OW%: Opponent’s Win Percentage = W(O)/G(O)
OOW%: Opponent’s Opponents’ Win Percentage = average of all 1 to 9 W(OO)/G(OO)


SOS: Strength of Schedule = 2/3 x OW% + 1/3 x OOW%
Source of stat: https://en.wikipedia.org/wiki/Strength_of_schedule
(BCS type because of lack of some data)
Explanation:
„How hard was your road?“
This stat describes how good the opponents you fought were. It takes into consideration how they did overall and how hard their road was to a somewhat lesser extend. This stat does not say anything about your performance, just analyzes how tough your enemies were along the road. In regular season, the SOS differed between .375 and .608, so there is quite the variety. Nonetheless, the fewer games you played, the less this stat does mean anything.

RPI: Relative Percent Index = 1/2 x OW% + 1/4 x OOW% + 1/4 W%
Source of stat: http://www.espn.com/nba/stats/rpi
Explanation:
“How well did you do considering your opponents?”
The RPI takes your own playing into account, so unlike the SOS, the RPI does tell you something about your performance. The problem here is: You could go 9-0 and would not even be Top10 because of the aforementioned variety. Also, more often than not, it is not your fault, you played “worse” player and if you are the best, none of your own opponents push this rating as high as you do it for them.

To overcome the problems the best players of every tier are naturally underrepresented in that stat, I tried his:
PER: Pokemon Efficiency Rating = RPI x {1 / [1 + (L/W)^1,83 ]}
Explanation:
“How did you do considering your opponents with more emphasis on your own performance?”
This stat is far from perfect and additionally “my creation” (yeah, I multiplied two numbers, sheesh), but it is needed because of the aforementioned problems. The RPI is multiplied with something that is called the “Expected Win Percentage”. Every sport follows this formula if you chose the Pythagorean method. The problem however, this exponent of 1.83 is obviously undefined for Pokemon (it varies from 1.83 for baseball to 16.5 for Basketball) and the stat usually does not take W-L, but runs, points, you name it… The lack of data forced me to use this standard exponent and the W-L-ratio, though. It is a skewed stat but in my opinion fixes the problem of underrepresentation of the “Tier winners”. The stat was normalized to compare it easier, which means the worst record corresponded to a clean 0 and the best record to 100. The non-normalized version is in the sheet as well, too.
So without further ado: ENJOY!

https://docs.google.com/spreadsheets/d/1wTSBfcoub8jWXLUqg11tT8amALZF6FgfiqzayZFQXvM/edit?usp=sharing

Shoutouts to all people who feed backed me yesterday. I found some mistakes, adjusted some numbers and I am still aware of the PER’s flaws. {1 / [1 + (L/W)1,83 ]} (aka Expected Winning Percentage) is in my opinion still the way to go with the exception of the 1,83 exponent, but there is nothing to find (at least for me) to get a better one. Feel free to suggest an alternative.
 
Last edited:
I have to say i thought this was incredible, I have never been a fan of just looking at W/L on a spreadsheet because i dont think it truly can show how good a player is or how much they bring to a team. This advanced stats could change the way managers draft in the future if used properly so I hope to see more of you around.
 

Mr.E

im the best
is a Pre-Contributoris a Past SPL Champion
I have to say i thought this was incredible, I have never been a fan of just looking at W/L on a spreadsheet because i dont think it truly can show how good a player is or how much they bring to a team. This advanced stats could change the way managers draft in the future if used properly so I hope to see more of you around.
It wouldn't, because managers are dumb anyway, but the sample size is too small to draw any meaningful conclusions. It's more for the fun of it.
 
It wouldn't, because managers are dumb anyway, but the sample size is too small to draw any meaningful conclusions. It's more for the fun of it.
Yeah about sample size... there are more tournaments and people join different ones... I wouldn‘t go as far as Sabella, but why shouldn‘t this help someone building a team?
 

Realistic Waters

Fake Tears
is a Tiering Contributor
rats, scooters were so close. GG ruiners :blobastonished:
I'm not gonna write an essay so i'll keep this short and sweet.

Huge thanks to Philip7086 and FlamingVictini for buying me and letting me play NU for the Scooters, I had a great time and I'm quite happy with my end result. lax and ict you guys are goats and it was great playing you, gl nerds. S/O NU community for the cheering and offering me so much support, I appreciate it so much (BEST TIER) . S/O to the rest of the Scooters for putting up with me even though I didn't talk much but you still showed your support giving me ideas and questioning my meme team choices hehe.

PEACE.
 
Whenever I see a new post in administrative decisions, I rush directly to see if a user or even a whole team gets banned, while also hoping for some drama in this thread. Today I saw a new post in the administrative decisions but i was disappointed to see just an announcement about our beloved harambes.

Anyway, the lack of new content in this thread should have given out the lack of such drama. The reason of this post is to check if there are people who do the same, and to also give those people the false impression that something actually happened :pimp:.
 
Whenever I see a new post in administrative decisions, I rush directly to see if a user or even a whole team gets banned, while also hoping for some drama in this thread. Today I saw a new post in the administrative decisions but i was disappointed to see just an announcement about our beloved harambes.

Anyway, the lack of new content in this thread should have given out the lack of such drama. The reason of this post is to check if there are people who do the same, and to also give those people the false impression that something actually happened :pimp:.
I feel you. Son of a b*tch.
 
Status
Not open for further replies.

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top