|
Fiddler's Green42661 Posts
The Process of Creating the Greatest of All Time List:
After reading most of the reactions, I thought I’d go into more detail on the step-by-step process of creating the lst. It’s actually pretty simple.
The first step was to take out criteria that I considered irrelevant in the making of the lists. Those criteria were:
Money won
Online Tournaments/Qualifiers
The reason is actually pretty simple. Lets say we take the argument that making the most money means therefore you had the most skill. In that case by that logic, you’d have to say Newbee are the greatest Dota2 players of all time (despite that they probably even weren’t the best Dota2 players of 2014). More relevantly, you’d have to tier every tournament solely based on prize earnings. Meaning something like Shoucraft NA run by TB is almost as important or more important than any MLG victory won in 2011. The same weekend MMA won MLG Columbus, Lyn won a Chinese SC2 tournament with 4 times the prize pool. Lyn was not 4 times the player MMA was at that moment. Fruitdealer’s GSL victory is more important than Life’s GSL victory in 2015. All money earning measures is how much money you earned. It’s a cool stat, but doesn’t particularly reflect on the skill of a tournament.
Secondly, online tournaments and qualifiers were discounted. ESL Korean qualifiers have a harder competitive list than WCS NA and WCS EU combined. But there is no prestige, no lan pressure, and the difference in reward between winning that qualifier and winning a tournament is massive. No one remembers who won what qualifier to make it into ESL in 2012, but everyone remembers Naniwa winning MLG in 2011 despite there being 0 Koreans in attendance (Do Select and Moonan even count?)
The second step was to quantify concepts to myself. Most importantly the difference between 1st and 2nd, and preparation formats vs non-prep formats.
The difference of course is the prestige, the difficulty, the reward, the legacy, the history. But it is also the difference of exactly one series. How many second places do you need in order to make it equivalent to a first? Some people say you could have an infinite amount. In that case You’d have to put Seed/Sniper/Jjakji over soO as an all time great. And I hope that looks as ludicrous to you as it does to me. Basically I settled this by making a mental note of first places and measuring the raods taken to the finals. For instance IEM Cologne 2014.
In terms of pure difficulty and skill I rate Polt’s runner-up of higher quality than HerO winning the exact same tournament. In the blink meta, it made it so only 3 Terrans became worth a damn in TvP: Maru, Taeja and Polt. HerO beat ForGG twice, Innovation, Jaedong and Polt. ForGG was already bad at TvP before that meta hit. Innovation got kicked out of the GSL because Zest used the blink all-ins to kick him out. Jaedong was good, but wasn’t nearly as impressive as he was in 2013. Polt on the other hand beat Mana, Stardust, Classic, Rain and then lost to HerO. Mana was pretty bad at PvT and couldn’t abuse the meta. Stardust was really good and was top 8 in WCS EU. Classic was cracking near Top 5 Protoss players of his race and would win the GSL a few months later. Rain was a Top 5 player of his race (arguably Top 2) and would be in the semi-finals of GSL in a month. HerO won in a favorable meta against Terran players weak against Blink All-ins, barely won against Jaedong, and then beat Polt. In his run I’d argue he only had two impressive performancs against Jaedong and Polt. Polt played in a god awful meta for Terran against one of the better Protoss’ of Europe in Stardust, beat a future GSL Champ, then beata Top 5- Top 2 Protoss before losing to HerO.
Basically for me it was all about context, the runs, the metas, who they played, how the played and the degrees therewithin.
Now preparation vs non-preparation. Basically I believe that preparation adds a large amount of complexity to any matchup. If given time, a player can devise and create a strategy that can maximize their chances of beating anyone, no matter how superior the other player is. It adds pressure and requires more skill and composure and hard work to the equation. Marathon tournaments have a completely different set of problems though. Foreign lans are much more susceptible to technical issues. Then you have to deal with jet lag, travel issues, being in a foreign country, fighting against a multitude of playstyles, you have to have endurance and you have to think of the fly. While I do respect the preparation format more, it is pretty ludicrious to say any non=preparation format is worse solely because of the format.
In the end, prep vs non-prep only mattered when compettiion at a foreign event was similar to a GSL (Tournaments like IPL 5, MLG Fall, IEM Toronto, IEM Cologne all spring to mind). In those cases, I added it into the back of my mind as a factor among many to consider.
Now I can get to the actual ranking part. First I thought up of every player I considered to be in the greatest of all time from 2010 to now. Then I listed all of their Top 4 placings at every Premier Event. Then I tiered each event they were at based on the competition. So then whats’ the difference between something like MarineKing’s MLG wins and Leenock’s MLG win? Why are they tiered differently. I basically went through the entire player list. I then wrote down how good they were for their respective race and only counted players that were either in the Top 10 or Top 5 of their race Then I cut everyone else out unless a player had an extremely hot run and played above their level (Sjow, Haypro, Patience all fit this example). I then took that list and theory crafted exactly how many players I’d need to add in order to say every one of the best players in the world at this very moment are here right now. Then I wrote down the paths of every single Top 4 run they did and only included players in my list. I then made sure to include players if a player was only a matchup god (Ryung in TvT is the first example here). So now I have a list of the greatest players and their achievements based on a 4 tier system on the competition that played there.
So for instance let’s take Parting’s WCS 2012 victory and measure it against MKP’s MLG Spring Championship top 3 placing. Parting’s WCS 2012 was a tier 3 tournament. It had basically all top 5 Protosses, 1 of the top 5 Zergs (Roro), a top 10 Zerg (Stephano) a middling Zerg in Curious and a bunch of strong foreign players that could on a very good day crack a top 10 performance of their race maybe (Grubby, Lucifron, Sen, Vortix, Scarlet). I’d need to add maybe 7 Koreans to make this a tier 1 event.
Now lets look at MLG Spring Championship. This was mid 2012, so MC, Alicia and Oz were all either top 5 or top 10 Protoss (this was during Alica’s run of second places, Oz coming off ro8 GSL, MC being MC). Then you had MKP, Polt, MMA, Thorzain, aLive. 2 of the Top 5 Terrans in MKP and Polt. MMA, Thorzain and aLive were around top 10. (MMA had come off Iron Squid and this was right as he fell. Thorzain DH Stockholm, aLive was falling from his top 2 Terran performance in early 2012, but was still in the upper echelon). Then you had 5 Zergs: DRG, Stephano, Violet, Symbol, Leenock. The only Zerg missing was Nestea (life had yet to exist). To knock this up to a tier one then you needed more Terrans (Byun, Mvp, Taeja) and at least 1 more of the Protosses (Naniwa, Genius or Seed).
Now look at their paths. Using my grading scale Parting only beat 4 notable opponents: Scarlett, Suppy (Who had a peak performance here), Sen and Creator. MKP to get 3rd beat: Symbol, Stephano, Thorzain, loss to DRG, loss to Alicia. If you really stretch here you could say Parting beat a top 10 Zerg, two foreigners in Suppy and Sen who played around top 10 level and Creator, Top 3 P. MKP beat a 2 Top 5 Zergs and then a top 10 Terran, and other multiple equivalent players to Sen/Suppy in Dream/Grubby/Golden. In terms of content you could make a case here that Marineking’s top 3 run was harder than Parting’s victory at WCS.
For me context was everything.
So basically after doing all of that I had a list of about 20 or so. I then measured their years of consistency. Let’s take Maru as an example. 3.5 years non-relevant. OSL to now he was a Top 3 Terran. During that Top 3 Terran era, he was top 1 for the first half of 2014 and in my eyes the best Terran of 2015 so far. I did this for every player.
I then annotated that by writing down their years as compared to the metas they played in. For instance, winning in a crapsack age as the blink era for Terran was worth more than winning in a golden age for Zerg in the bl/infestor era.
Then I judged Innovation and refinement. What builds/styles did a player innovate (in the sense that he was the first to do a build consistently, players credited that player with the build and/or players copied him after he had done it - like Stephano and the 3 hatch) Then there was refinement. Nestea created and refined Muta ling/bling, but DRG took that refinement a step further than that. Generally I had to measure the degrees of refinement and innovation. Let’s take for example Life. He created a strong counter-attack style that was extremely refined and was uncopyable for 2012. While never replicated perfectly, it was then used by multiple players from 2013 onwards (most notably Curious, Leenock, Jaedong, Soulkey and Pet). He got maximum points possible in refinement and a decent amount of points in innovation. Stephano got maximum points in both.
Then you had to measure adversity, which I touched upon. Basically this comes down to metas, how many players of your race could you learn from during your time (For instance Nestea was the only zerg worth a damn in 2011 until DRG/Leenock rose up), outside influences (MMA’s team situation) or injuries.
Then I measured things like relative increased skill overtime, how dropout/retirements may have affected the overall skill gap, and then measured the differential between TOp 1 and Top 5 of each race. For instance in the latter half of 2012, Top 3 Protoss were: Parting, Creator, Rain. It then went on a pretty decent gap between them and 4th pce of P. In 2013 It went Soulkey, Jaedong and then a humungous gap between them and 3rd best Zerg. 2014, it went soO and then a gap so massive it was only comparable to Nestea’s 2011 era, until you got to the second best Zerg.
Then I added in X factors for each player: Clutch, composure, versatility, play style, series planning.
Now take every factor I measured up and then grade two players head-to-head on every level (measured and balanced runs, prestige, achievements, time of consistency, peak consistency, innovation, refinement, adversity and x factors) and grade them. That’s basically what I did for every player that could have made the list of the Greatest of All Time. If you go through the entire thing, you could probably deduce the rest of my Top 10. Simple right?
 
|
Sounds it 
How long did this process take you?
|
wow. more detailed than I thought it would. I'd ask you questions but I'm afraid you've aleready answered them and I'm just failing at reading comprehension
|
Fiddler's Green42661 Posts
About 20 hours of research. I basically had to chart a line of every player's careers (and matchups) to know exactly how strong a run was. Then the infinite comparing, then I went around and asked ppl who may have had better perspective on certain players. This would have taken a lot longer if I hadn't been studying SC2 for years already.
|
That's mightily impressive. And yeah if I were to try and do the same it would take me at least triple that time (and likely a lot longer if I started watching replays. haha). I only play intermittently but love watching the GSL so I totally dig all the articles you guys produce. Keen as for parts 2&3!
|
If I were a girl, this kind of detail would make me moist.
Spreadsheets upon spreadsheets
|
uh congrats on 31115 posts blog?
no seriously thank you for writing it all out it makes things extremely clear and I appreciate it. Using this we prob could figure out the exact list but that would take too much effort
|
"pretty simple"
Bloody hell.
|
Magnificent. That's such a good explanation that I'll even wait until the whole list is published before disagreeing further .
|
1001 YEARS KESPAJAIL22272 Posts
it is in our best interest if you keep disagreeing with everything ever
|
United States97276 Posts
is stephano in the top 10?
|
no mention of squirtle; squirtle and mc were leagues ahead of all the other toss when it was considered the worst race they were winning everything. No mention of bomber either, the terran winning stuff besides mvp when terrans were struggling against the super zergs
|
Which database did you use to remember all this though? TLPD or Aligulac??? :-D
(Of course you spent hours upon hours on Liquipedia, as do we all)
|
The fact alone that you judge people based on "the meta", which is mostly a thing that exists only in the minds of TL posters and whiny foreigners makes your ranking irrelevant. It's almost sad how much work and passion you have invested into it only to become only an extremely sophisticated version of "I like those players the most". Your "tiering" process has the same flaw as almost everything ever done on TL - it includes players who are relevant only because of their success outside Korea and taking them into account increases the relevance of the foreign tourneys in a circular argument.
But one should not be just negative. I am not willing to put in 20 hours, so the example will not be as refined, but let's see just the rough outlines how would I go about such list in simple bullet points:
- open the list of 1st/2nd places in GSL (excludng "special tournaments" because they are weird), OSL, SSL
- identify players that are on this list at least twice. These are people that show a minimal required level of consistence in the only tournaments that actually matter:
Nestea, MC, Mvp, Life, Innovation, soO, Rain, Maru, MKP,DRG
- sigh with a slight relief that the list has 10 names so can be used without pulling other names from ass
- sigh with a slight disgust that I have crafted the criteria so stupidly that MKP is on the list and Parting isn't
- stare into the liquipedia pages endlessly to find more people that would push MKP and DRG out from top 10
- contemplate including 2014 "GSL global champs" to gain Parting but have to endure including Zest and lose all credibility
- contemplate how everyone is so crazy about sOs who did not reach the finals even once, appreciate that Stuchiu put him 16th
- be happy that Taeja is not on the list, even though the method was chosen primarily with this goal in mind
- still contemplate how badly I want Parting to be on the list, why doesn't the greedy bastard win a GSL instead of collecting easy money around the world?
Now it is getting hard and random:
- assign points for the placements
- 2 points 1st place
- 1 point 2nd place
- 1 point for any extra year your greatness spans (have a point from 2014? get a point if you have a point from 2012, get four if you were so good in 2010) - it is supposed to be of all times, right? (i call this "the lifeline ")
Nestea 7
MC 7
Mvp 8 (yay!)
Life 7
Innovation 4
soO 5
Rain 4
Maru 6
MKP 3
DRG 3
- be honestly surprised that the out-of-the-ass point system at least puts MKP and DRG last and relieved that it indeed puts Mvp first (after being a little scared by the discovery that he has only two different years)
- see the need to decide tie-breakers - as it is "all times", tie-breakers are decided by length between points (guess why)
- assemble the final list in the correct order:
Mvp
Life
Mc
Nestea
Maru
soO
Innovation
Rain
Parting
DRG
|
Austria24422 Posts
No ranking is irrelevant. Calling a subjective ranking irrelevant makes your post irrelevant.
|
On April 14 2015 17:36 opisska wrote:+ Show Spoiler +The fact alone that you judge people based on "the meta", which is mostly a thing that exists only in the minds of TL posters and whiny foreigners makes your ranking irrelevant. It's almost sad how much work and passion you have invested into it only to become only an extremely sophisticated version of "I like those players the most". Your "tiering" process has the same flaw as almost everything ever done on TL - it includes players who are relevant only because of their success outside Korea and taking them into account increases the relevance of the foreign tourneys in a circular argument.
But one should not be just negative. I am not willing to put in 20 hours, so the example will not be as refined, but let's see just the rough outlines how would I go about such list in simple bullet points:
- open the list of 1st/2nd places in GSL (excludng "special tournaments" because they are weird), OSL, SSL
- identify players that are on this list at least twice. These are people that show a minimal required level of consistence in the only tournaments that actually matter:
Nestea, MC, Mvp, Life, Innovation, soO, Rain, Maru, MKP,DRG
- sigh with a slight relief that the list has 10 names so can be used without pulling other names from ass
- sigh with a slight disgust that I have crafted the criteria so stupidly that MKP is on the list and Parting isn't
- stare into the liquipedia pages endlessly to find more people that would push MKP and DRG out from top 10
- contemplate including 2014 "GSL global champs" to gain Parting but have to endure including Zest and lose all credibility
- contemplate how everyone is so crazy about sOs who did not reach the finals even once, appreciate that Stuchiu put him 16th
- be happy that Taeja is not on the list, even though the method was chosen primarily with this goal in mind
- still contemplate how badly I want Parting to be on the list, why doesn't the greedy bastard win a GSL instead of collecting easy money around the world?
Now it is getting hard and random:
- assign points for the placements
- 2 points 1st place
- 1 point 2nd place
- 1 point for any extra year your greatness spans (have a point from 2014? get a point if you have a point from 2012, get four if you were so good in 2010) - it is supposed to be of all times, right? (i call this "the lifeline ")
Nestea 7
MC 7
Mvp 8 (yay!)
Life 7
Innovation 4
soO 5
Rain 4
Maru 6
MKP 3
DRG 3
- be honestly surprised that the out-of-the-ass point system at least puts MKP and DRG last and relieved that it indeed puts Mvp first (after being a little scared by the discovery that he has only two different years)
- see the need to decide tie-breakers - as it is "all times", tie-breakers are decided by length between points (guess why)
- assemble the final list in the correct order:
Mvp
Life
Mc
Nestea
Maru
soO
Innovation
Rain
Parting
DRG
Sry but excluding all the foreign tournaments and korean special events "cause they are weird" is completely ridiculous.
|
On April 14 2015 20:25 The_Red_Viper wrote:Show nested quote +On April 14 2015 17:36 opisska wrote:+ Show Spoiler +The fact alone that you judge people based on "the meta", which is mostly a thing that exists only in the minds of TL posters and whiny foreigners makes your ranking irrelevant. It's almost sad how much work and passion you have invested into it only to become only an extremely sophisticated version of "I like those players the most". Your "tiering" process has the same flaw as almost everything ever done on TL - it includes players who are relevant only because of their success outside Korea and taking them into account increases the relevance of the foreign tourneys in a circular argument.
But one should not be just negative. I am not willing to put in 20 hours, so the example will not be as refined, but let's see just the rough outlines how would I go about such list in simple bullet points:
- open the list of 1st/2nd places in GSL (excludng "special tournaments" because they are weird), OSL, SSL
- identify players that are on this list at least twice. These are people that show a minimal required level of consistence in the only tournaments that actually matter:
Nestea, MC, Mvp, Life, Innovation, soO, Rain, Maru, MKP,DRG
- sigh with a slight relief that the list has 10 names so can be used without pulling other names from ass
- sigh with a slight disgust that I have crafted the criteria so stupidly that MKP is on the list and Parting isn't
- stare into the liquipedia pages endlessly to find more people that would push MKP and DRG out from top 10
- contemplate including 2014 "GSL global champs" to gain Parting but have to endure including Zest and lose all credibility
- contemplate how everyone is so crazy about sOs who did not reach the finals even once, appreciate that Stuchiu put him 16th
- be happy that Taeja is not on the list, even though the method was chosen primarily with this goal in mind
- still contemplate how badly I want Parting to be on the list, why doesn't the greedy bastard win a GSL instead of collecting easy money around the world?
Now it is getting hard and random:
- assign points for the placements
- 2 points 1st place
- 1 point 2nd place
- 1 point for any extra year your greatness spans (have a point from 2014? get a point if you have a point from 2012, get four if you were so good in 2010) - it is supposed to be of all times, right? (i call this "the lifeline ")
Nestea 7
MC 7
Mvp 8 (yay!)
Life 7
Innovation 4
soO 5
Rain 4
Maru 6
MKP 3
DRG 3
- be honestly surprised that the out-of-the-ass point system at least puts MKP and DRG last and relieved that it indeed puts Mvp first (after being a little scared by the discovery that he has only two different years)
- see the need to decide tie-breakers - as it is "all times", tie-breakers are decided by length between points (guess why)
- assemble the final list in the correct order:
Mvp
Life
Mc
Nestea
Maru
soO
Innovation
Rain
Parting
DRG
Sry but excluding all the foreign tournaments and korean special events "cause they are weird" is completely ridiculous.
The process described is not supposed to be rocket science and is not supposed to be taken dead seriously, I hope that is clear from the post.
However, what I really mean seriously is that if I am to call someone "one of the greatest of all time", I require said player to do well in the best competition in the world and that is the Korean leagues. From these, the regular GSL and recently the OSLs and SSL have a clear, readable structure guaranteeing the highest possible level of competition, while the "special" events have the tendency to have players from a more limited pool, sometimes even invites based on foreign tourneys etc. In general, judging their relevance would require going through the players one by one and that is exactly the thing I think should not be done, because it introduces bias inevitably.
As for foreign tournaments, I am not disregarding them "because they are weird" (that was refering to the special korean leagues with the reasoning above behind it), but because I do not consider them the highest level of competition. You can go on forever naming "strong" players in a given tourney, but as I have already explained, a lot of these are considered "strong" because they do well in foreigner tournaments and that is an argument by circle. Yes, many of them are GSL players or even title holders, but seeing as the player list of the GSL itself is in a constant flux, that by itself is not guarantee of the highest level of competition.
At the end of the day, the strongest argument for me not to value the foreigner competition as highly as many do here is the observation that a majority of successful players of the foreigner circuit do indeed fail any time they try to challenge themselves with the GSL. Yes, there are exceptions (lately MMA for example) but these are much more exceptions that rule.
edit: however let us see what the Stuchiu's list actually is. I am going to base any further aggression against the list largely on the position awarded to Taeja.
|
Well i agree that the korean tournaments should be the most important factor. I simply didn't agree with the complete lack of foreign tournaments (while i see what you mean with circular reasoning, i don't think this is entirely true here) and korean special events.
They should be less important (gsl/ssl/osl > korean special (hot6ix,kespa cup) > foreign tournaments (and these should be rated differently too, a HSC =/= an IEM) ; in general, there are exceptions)
So yeah i kinda agree with you but i guess i didn't notice that you aren't 100% serious 
|
On April 14 2015 20:54 The_Red_Viper wrote:Well i agree that the korean tournaments should be the most important factor. I simply didn't agree with the complete lack of foreign tournaments (while i see what you mean with circular reasoning, i don't think this is entirely true here) and korean special events. They should be less important (gsl/ssl/osl > korean special (hot6ix,kespa cup) > foreign tournaments (and these should be rated differently too, a HSC =/= an IEM) ; in general, there are exceptions) So yeah i kinda agree with you but i guess i didn't notice that you aren't 100% serious
Now I had to consider that people may actually take everything I write here seriously. That considered, it is likely that some people think that I am a total moron 
|
On April 14 2015 21:10 opisska wrote:Show nested quote +On April 14 2015 20:54 The_Red_Viper wrote:Well i agree that the korean tournaments should be the most important factor. I simply didn't agree with the complete lack of foreign tournaments (while i see what you mean with circular reasoning, i don't think this is entirely true here) and korean special events. They should be less important (gsl/ssl/osl > korean special (hot6ix,kespa cup) > foreign tournaments (and these should be rated differently too, a HSC =/= an IEM) ; in general, there are exceptions) So yeah i kinda agree with you but i guess i didn't notice that you aren't 100% serious Now I had to consider that people may actually take everything I write here seriously. That considered, it is likely that some people think that I am a total moron
Haha i don't know, i just disagreed with this comment :D
But yeah i think Taeja will be way too high as well ^^
|
|
|
|
|
|
EPT
