0:00 – Intro
0:27 – What is the GBL Algorithm
2:40 – My GBL Stats
5:00 – Reddit Post: Why 50% Win Rate
6:05 – Counter Argument to 50% Win Rate (Not in Financial Interest of Niantic)
7:55 – Statistics (Standard Distribution Graph of Matches)
11:00 – Putting a Number of Hard Counter Games
12:55 – Confirmation Bias
14:00 – PokeDaxi Legend Run
15:00 – I am a Teaching Channel to Help you Hit Legend
15:45 – 2 Things Stopping You from Hitting Legend
18:45 – Final Thoughts/My Opinion
GO Battle League Timeless Travel Information: https://pokemongolive.com/en/post/go-battle-league-timeless-travels/
My Referral Code for Newer Players: RVGV4MJRC
Submit battles here: https://docs.google.com/forms/d/e/1FAIpQLScKj-Td6kLsFEcKg17eHsuGjS-oPygYMOmxekxpItNwiSaYxw/viewform
The best Ottawa Pokemon Go Youtuber….is Poke Daxi but I am here too
This channel is where you come if you want ideas for:
Great League Teams
Ultra League Teams
Master League Teams
Beyond the Hype. I created this series as there is a lot of click bait and false impressions of pokemon and I want to give you a real in depth analysis of whether a pokemon is worth spending dust and candies on.
The new season of Timeless Travel has been announced with buffs to pokemon like talonflame, toxicroak, and azumarill and nerfs to steelix breaking swipe and medicham. Here is a full analysis of the new buffs, nerfs, and pokemon with new moves.
In this video we are talking about the GBL Algorithm. The GBL Algorithm, or matchmaking system, is a theory that the GO Battle League matchmaking system is aimed at keeping you at a 50% win rate and purposely hard countering your teams to keep you there.
26 Comments
There ain't no algorithm, and my longest streak is 42. I tested this by tanking. My highest was Veteran at 2700 so you could probably get a hold of my overall skill level in this game. Before Go Battle Weekend last season, I tanked every set the whole week so I reach 600 ELO and then test out the myth. If there is an algorithm that hard counters your team every time you win too much, then I'd probably get bad lead into bad lead into bad lead. But that wasn't the case. I probably had like 5 bad leads in total and my medi punched through (literally) every single kecleon, wartortle, and every other random stuff you find down there. So it's not about an algorithm pitching you against your worst matchup possible, but people are just better the higher you go.
So just like Dan said, if you ain't hitting legend, it's a skill issue. Suck it in, and keep grinding. <– this last part was a message for myself btw sobs
DanOttawa POGO never fails to make good content đ đ
Every game that utilizes an ELO system ends up with a community of conspiracy theory people that insist there is secret code that matchmakes based on punishing the individual player. People are bad at statistics. League of Legends has a decade+ of conspiracy theories about "ELO hell", "losers queue", etc.
thank you
Another interesting thing here is that the games use some variant of the ELO system. This ranking method was designed specifically for ranking the game of chess. One of the key features of the ELO system is to tune your rank overtime so that you are matched with player of even skill. The intent is to produce a roughly 50% win rate for the average player. You can see this in GBL if you wait a month to start grinding. If you are a 2200 player you will find that you go on a massive streake during the first 20 levels if you wait for the other good players to rank up. Once you approach 2200 you will being to regress to the intended 50% win rate. That said, ELO is a less than desirable ranking system for GBL because of the presence of weaknesses and resistances in pokemon combat. One thing that is needed for ELO to function is an assumption that at the beginning of a given match each player is on a even field. This is true of chess. No matter the ELO of the players at the start of the game no player has an advantage. This is not true for GBL matches. To make a long story short, any ranking system using ELO is designed to reduce variance in skill match ups so when working properly with all the needed assumptions met you will know you are at the correct rating when you have roughly a 50% win rate against players of similar ELO.
This video made me feel good about GBL
great! now how am i supposed to cope ! !
I remember one season where I was in the upside down world where Iâd win my bad leads and then lose my good ones.
I like the discussion of this video. I'm not sure what I believe when it comes to the GBL algorithm, but I've definitely had my fair share of frustrating matches. The days where you change your team 5 times in a set and each match you get hard countered, or you see a bunch of one pokemon in one set, so you put something in to hard counter it, then suddenly that pokemon is no where to be seen. Those games are incredibly annoying. My win % is roughly 52% though and I usually climb every season, so it's clear its not a perfect set 50%.
Sounds like an accurate assessment to me. Great vid
Thank you! I always roll my eyes when other YouTubers talk about this – confirmation bias and superiority bias all over the place
Please send this to PokeAk he's insufferable
Honestly I really love what you said about confirmation bias. Itâs easy to believe that youâre being specifically targeted and thatâs why youâre on a losing streak but you conveniently forget that there were just as many times that you completely hard countered your opponent.
People put way too much stock into lead matchups – I mean yeah theyâre important but having a bad lead isnât an auto loss just as having a good lead isnât an auto win. What both players have in the back is just as important and can swing the match just as much as the lead matchup. Just as an example, Iâm currently running a Poliwrath/Dragonair/Skeledirge line and earlier on faced a Lanturn/Dewgong/Trevenant line. Hard countered on the lead and on switch but still won because I managed to align Skeledirge onto Trev in the back during the endgame.
Having the skill set, knowledge of your team and experience to play around hard counters is what seperates the best players from the rest of us – sure thereâs always going to be games where you just get completely countered and there is no win con at all, it happens. It doesnât mean thereâs some hidden system deliberately keeping you down – if youâre running a meta team youâre going to run into hard counters from time to time, there is no âperfectâ team that has absolutely no weakness whatsoever.
Why people seem to believe that Niantic have the competence to implement a system that matchmakes thousands of battles every minute right down to each individual players specific Pokemon and movesets and organise each pairing into âhard counteredâ vs âhard counteringâ when they canât even keep the game itself running properly most of the time is beyond me.
Itâs also worth noting with regards to the whole â50% win/loss ratioâ thing that say for example you go on a winning streak and you climb 100-200 elo, you need to take into account that at that point youâre going to be playing more skilled opponents than you were before so youâre naturally going to start losing more games again which of course is going to even out that W/L ratio.
I am sorry but I dissagree bro, I do not get hard contered 300 + times a seoson, its a lot more thane that. I have battleded in 28 600 battles and won 14 400 and I can say that there is deffienetly a system that mathces you with either a hard counter in the lead or specific moves that are just the right moves against your team. Multiple times I have lead with lets say Tina origins in ML and I get 3 straigh dragon leads, I then switch to a togekiss lead and the next two leads are Mewtwo and Melmetal…. I have even made a demonstration to my girlfriend last time I swichem my lead in GL to something weak to Ice and I went into a battle telling my GF look now I will get a Walrein against me…. and booom Walrein.
Also, I dissagree that Niantic does not make money from GBL and that they dont have any reason to have an algorithm. You can use Battle passes, which are paid for better rewards, so there is a paid item you use in GBL and the also do not want you to get only good sets – 4/5 5/5 because you will get a bunch more dust and rare candies, which you can then feed to legendaries, which they want you to raid for those candy instead.
On the subject od why people are winning and getting to legend, it has nothing to do with your record, you yourself have a 50% win rate for last season and you hit legend. Its about making a run and a string of good sets when it counts and most people can do that due to skill, counting moves, catching moves, being able to take advantage of an opponents mistake, but alot of people cant and dont want do to that. Every season I get to Ace or Veteran and thats it, and I am OK I dont care about hitting legend, I want to go in GBL get my sets for the day, get my dust, some rare cany an encounter maybe and it is incredibly frustrating when you are hard countered all the time. For me, it feels like a chore everyday, and not something I do because I find it fun. But I continue to battle because the rewards are still good and I cannot go out and grind everyday.
7:44 Thats in accurate: on the surface this argument might seem very logical. yet if you stop and think deeper, you realize that building Teams for GBL with XL Monst to be more specific, you drain your Stardust resource so much, that you are pushed to do these tasks like egg hatching raiding and go outside. By simply Only doing Battles you wont keep up with the rising Stardust cost as new mons get into the meta etc. And dont Forgett, Master League is Raid Mons League aka. Legendarys aka. a lot of Stardust investments and Candys.
For the arguments sake: Ppl at the highest of leader boards understand to force plays the opponent Has to react perfectly to keep their lead even if the matchup is in the favor of the opponent. ( same in chess, there its called '' Only Moves '' ) and as you know, some times you just can not win. đ
you bet your ass i dived into the abyss for that ! x_DD
Nvm saw pokeak video on you, hes a bozo
Hi nice video and topic. i have a question related to the state of the game now. is still laggy as it was last season. i was discusted at the end i stoped trying.
I know there's no algorithm, but I'll damned if I don't blame it every time I lose.
Last 2 seasons I reached vet, this season I seem to be stuck on level 20, each one I've had a 51% win rate.
I think itâs interesting to have a game, where there is no financial gain for you as the player, to be this polarizing. Itâs a free game and itâs your choice to spend money on it if you want. But people get so mad, angry, and upset about so much with it. Why? How much money, coins, or dust do you get for hitting legend? If you start treating a game like a job, thatâs when this conversation comes into play I feel. My ultimate response to the âis there an algorithm questionâ is who cares? Dan is right that if you want to hit legend, just like anything else, it takes time and desire to learn the inâs and out. Granted some people are just naturally able to pick this up but when you listen to the top players talk about the math, timing, percentages, ask yourself (before blaming the algorithm) if you knew all that. I play for fun and watch these videos because I want to learn more and enjoy the game.
It's interesting to watch today's Poke AK video, that seems to address you without addressing you. "those who do not believe in it […] have yet to prove that it's not real. Every time they try to come up with some sort of pie chart, or graph, or I got this degree, I got that degree, I know what I'm takking aboot it always falls flat and the comments are always going to reflect that" […] "there is an algo, and you know it, and you can't stop talking about it, and they know it's gonna bring in their views" […] "you got to love it, the cope… it's amazing". Also they show their total battle stats in the same way you do.
I think it's relatively easy to prove that there is an algorithm but it cannot be done with only one account: you need at least two accounts, with two different teams, A and B, as opposite as possible, with similar ELO. You do as much battles as you can with both, and note every team you encounter, using for example gobattlelog. If there is an algorithm, A and B will encounter different metas. For statistical significance, a lot of accounts should be involved, and split in two groups A and B running team A and team B. I guess that would require silph-level collaboration.
So, in terms of ELO; do you get more points if more of your team survives to the end of the round?
So that sounds like similar SBMM that call of duty uses to fully manipulate the game and leaves your skill where they decide. Unacceptable
I stopped playing GBL for the purpose of hitting legend cos I (for now) can't play consistently or frequently enough to turn those 381 times I get hard counter to my favour lol. Wish I could cos I've learned a lot from the likes of you and Jonkus but…meh some other time lol
This is a spicy video! I do notice though that a lot of times sets swing wildly 1-4 to 4-1, you would probably have to be like top player level to flip a lot of those games. I tried out bronzong today and I faced a lot of gfisk trev teams (rubix masters team shown on itsaxn's stream yesterday), and then after losing a bunch I started winning including a team with aba poison nidoqueen and clodsire, I don't know if thats coincidence but it seems extremely rare
Truer words couldn't have been spoken. I frequently fall into the trap that because a content creator's team got them this insane streak or got them Legend that it would happen to me, too. Recently I copied House Stark's team and then Ilqm's team. Performed poorly. Went with a team I knew well and positive sets the rest of the day.