|
On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept.
I was nodding my head along to this, thinking, the latter is indeed greater, lol.
Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day.
|
On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day.
I am not quite sure if I understand the example.
Is it:
Both alternatives include same tournament count and same time span.
Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th.
Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some.
Is that the example?
If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4.
Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4.
Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing.
Can we deduce from the described alternatives 1 and 2 who is the better player?
Only with context:
- why are there always the same 4 on top in alternative 1?
- Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)?
- is the player pool equally big?
- for the different metrics:
tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money)
Efficiency score as well, as it is depending on tournament score
Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals
tournament win %, yes of course, as alternative 2 won less for the same tournament count.
Average placement as well.
Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized.
Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral.
|
Russian Federation40190 Posts
On June 12 2025 23:11 PremoBeats wrote:+ Show Spoiler +On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day. I am not quite sure if I understand the example. Is it: Both alternatives include same tournament count and same time span. Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th. Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some. Is that the example? If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4. Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4. Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing. Can we deduce from the described alternatives 1 and 2 who is the better player? Only with context: - why are there always the same 4 on top in alternative 1? - Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)? - is the player pool equally big? - for the different metrics: tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money) Efficiency score as well, as it is depending on tournament score Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals tournament win %, yes of course, as alternative 2 won less for the same tournament count. Average placement as well. Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized. Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral.
Alternative 1 is on point (though i'd rather use Clem as example), alternative 2 is almost there but instead of having anyone else show up consistently everyone else either straight up does not repeat or at most appears 2 times out of 10 tournaments in top 4.
To ground it in real albeit very mean example: would soO be GOAT if his Kong streak didn't phase him and he would go on to place second in every big tournament for a few more years until retirement/military service?
The point here is that no matter what criteria you use, if alternative 1 is favoured, it is likely you are measuring dominance but not exactly greatness.
|
On June 13 2025 05:38 lolfail9001 wrote:Show nested quote +On June 12 2025 23:11 PremoBeats wrote:+ Show Spoiler +On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day. I am not quite sure if I understand the example. Is it: Both alternatives include same tournament count and same time span. Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th. Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some. Is that the example? If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4. Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4. Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing. Can we deduce from the described alternatives 1 and 2 who is the better player? Only with context: - why are there always the same 4 on top in alternative 1? - Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)? - is the player pool equally big? - for the different metrics: tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money) Efficiency score as well, as it is depending on tournament score Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals tournament win %, yes of course, as alternative 2 won less for the same tournament count. Average placement as well. Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized. Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral. Alternative 1 is on point (though i'd rather use Clem as example), alternative 2 is almost there but instead of having anyone else show up consistently everyone else either straight up does not repeat or at most appears 2 times out of 10 tournaments in top 4. To ground it in real albeit very mean example: would soO be GOAT if his Kong streak didn't phase him and he would go on to place second in every big tournament for a few more years until retirement/military service? The point here is that no matter what criteria you use, if alternative 1 is favoured, it is likely you are measuring dominance but not exactly greatness.
There is much to unpack.
In my opinion there is no dominance OR greatness. As I see it, greatness is compromised of hard numbers as well as context based, harder to grasp phenomena.
Dominance, as well as efficiency, longevity and consistency are qualities that can be measured and add up to greatness.
The strength of competition or the amount of tournaments that one is able to beat (or if a player pool is dispersed among many tournmanets) all are contextual perspectives that need to be taken into account.
Dominance in my opinion is necessary but not sufficient for greatness, thus your critique is valid as a thought experiment, but I don't see any issue with my overall analysis.
My methodology - as all sports metric do - leans towards dominance but I've clearly accounted for context (and I would say more so than most other analyses). Contextualized dominance, consistency, efficiency and duration are the most important metrics, although one can discuss their weighting perfectly fine.
It is a little bit like the discussion if slump-/non prime years should be disregarded. Some users argue that it is not fair that people's long careers are negatively awarded in efficiency or tournament win %, although these are specifically metrics designed to check for these criteria and compare players in them.
In my opinion, only looking at prime years is unfair to players who were able to distinguish themselves from others, by playing super efficient for several years, but one needs to take into account that efficiency is easier to achieve in short(er) careers.
That is the best approach to look at this whole thing: Get the hard numbers, then add context to them.
|
United States1886 Posts
On June 13 2025 05:38 lolfail9001 wrote:Show nested quote +On June 12 2025 23:11 PremoBeats wrote:+ Show Spoiler +On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day. I am not quite sure if I understand the example. Is it: Both alternatives include same tournament count and same time span. Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th. Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some. Is that the example? If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4. Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4. Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing. Can we deduce from the described alternatives 1 and 2 who is the better player? Only with context: - why are there always the same 4 on top in alternative 1? - Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)? - is the player pool equally big? - for the different metrics: tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money) Efficiency score as well, as it is depending on tournament score Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals tournament win %, yes of course, as alternative 2 won less for the same tournament count. Average placement as well. Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized. Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral. Alternative 1 is on point (though i'd rather use Clem as example), alternative 2 is almost there but instead of having anyone else show up consistently everyone else either straight up does not repeat or at most appears 2 times out of 10 tournaments in top 4. To ground it in real albeit very mean example: would soO be GOAT if his Kong streak didn't phase him and he would go on to place second in every big tournament for a few more years until retirement/military service? The point here is that no matter what criteria you use, if alternative 1 is favoured, it is likely you are measuring dominance but not exactly greatness.
I dunno what you mean by "kong streak phased him". soO actually went 4-0 in the finals of the four premier/major events he reached from 2019 onwards. He won IEM Katowice 2019, Master's Coliseum 4 and 5 and TSL 5 in 2020. soO's issue was that he was quite poor in the late game and, as SC2 aged and games began to reach said stage at an earlier and earlier point, his results began to suffer more and more.
|
On June 13 2025 15:09 Mizenhauer wrote:Show nested quote +On June 13 2025 05:38 lolfail9001 wrote:On June 12 2025 23:11 PremoBeats wrote:+ Show Spoiler +On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day. I am not quite sure if I understand the example. Is it: Both alternatives include same tournament count and same time span. Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th. Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some. Is that the example? If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4. Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4. Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing. Can we deduce from the described alternatives 1 and 2 who is the better player? Only with context: - why are there always the same 4 on top in alternative 1? - Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)? - is the player pool equally big? - for the different metrics: tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money) Efficiency score as well, as it is depending on tournament score Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals tournament win %, yes of course, as alternative 2 won less for the same tournament count. Average placement as well. Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized. Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral. Alternative 1 is on point (though i'd rather use Clem as example), alternative 2 is almost there but instead of having anyone else show up consistently everyone else either straight up does not repeat or at most appears 2 times out of 10 tournaments in top 4. To ground it in real albeit very mean example: would soO be GOAT if his Kong streak didn't phase him and he would go on to place second in every big tournament for a few more years until retirement/military service? The point here is that no matter what criteria you use, if alternative 1 is favoured, it is likely you are measuring dominance but not exactly greatness. I dunno what you mean by "kong streak phased him". soO actually went 4-0 in the finals of the four premier/major events he reached from 2019 onwards. He won IEM Katowice 2019, Master's Coliseum 4 and 5 and TSL 5 in 2020. soO's issue was that he was quite poor in the late game and, as SC2 aged and games began to reach said stage at an earlier and earlier point, his results began to suffer more and more.
Also we should remember when soO had that GSL kong streak in HOTS zerg wasn't as strong as it is in LOTV. I think people underrestimate how great soO was in 2014.
|
United States1886 Posts
On June 13 2025 15:31 dedede wrote:Show nested quote +On June 13 2025 15:09 Mizenhauer wrote:On June 13 2025 05:38 lolfail9001 wrote:On June 12 2025 23:11 PremoBeats wrote:+ Show Spoiler +On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day. I am not quite sure if I understand the example. Is it: Both alternatives include same tournament count and same time span. Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th. Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some. Is that the example? If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4. Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4. Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing. Can we deduce from the described alternatives 1 and 2 who is the better player? Only with context: - why are there always the same 4 on top in alternative 1? - Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)? - is the player pool equally big? - for the different metrics: tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money) Efficiency score as well, as it is depending on tournament score Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals tournament win %, yes of course, as alternative 2 won less for the same tournament count. Average placement as well. Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized. Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral. Alternative 1 is on point (though i'd rather use Clem as example), alternative 2 is almost there but instead of having anyone else show up consistently everyone else either straight up does not repeat or at most appears 2 times out of 10 tournaments in top 4. To ground it in real albeit very mean example: would soO be GOAT if his Kong streak didn't phase him and he would go on to place second in every big tournament for a few more years until retirement/military service? The point here is that no matter what criteria you use, if alternative 1 is favoured, it is likely you are measuring dominance but not exactly greatness. I dunno what you mean by "kong streak phased him". soO actually went 4-0 in the finals of the four premier/major events he reached from 2019 onwards. He won IEM Katowice 2019, Master's Coliseum 4 and 5 and TSL 5 in 2020. soO's issue was that he was quite poor in the late game and, as SC2 aged and games began to reach said stage at an earlier and earlier point, his results began to suffer more and more. Also we should remember when soO had that GSL kong streak in HOTS zerg wasn't as strong as it is in LOTV. I think people underrestimate how great soO was in 2014.
Way to make this about Zerg being OP in Lotv!
|
On June 13 2025 17:32 Mizenhauer wrote:Show nested quote +On June 13 2025 15:31 dedede wrote:On June 13 2025 15:09 Mizenhauer wrote:On June 13 2025 05:38 lolfail9001 wrote:On June 12 2025 23:11 PremoBeats wrote:+ Show Spoiler +On June 12 2025 21:01 ejozl wrote:Show nested quote +On June 02 2025 22:52 lolfail9001 wrote:
Frankly speaking, others already pointed out, but the second a counter factual of someone not qualifying for events for a few years potentially improving their overall standing in "GOAT" analysis comes up, it screams that one messed up their methodology from the ground up and should re-do it again. There a few other sanity checks like that i have in mind: for example is someone winning every tournament with the same top 4 placements for a year or two more or less "great" than someone consistently top 4 in tournaments that otherwise have completely different top 4 in every single one in the same time span (and for check's sake, same tournament count). By OP's metrics i am fairly sure the former is "great"er on every single measurement but we all know who would be the GOAT among the two (i am not stating those are real players, so don't let possible associations with real players confuse you).
EDIT: I'd figure the former is indeed more dominant but "greatness" is far more ephemeral of concept. I was nodding my head along to this, thinking, the latter is indeed greater, lol. Also, maru just won and is he even playing 2 hours a day, who knows. Compare that to pros killing themselves playing 12 hours a day. I am not quite sure if I understand the example. Is it: Both alternatives include same tournament count and same time span. Alternative 1: Maru is winning every tournament and Rogue, Cure, Zoun are always the placed 2nd to 4th. Alternative 2: herO places top 4 in tournaments but on top of Rogue, Cure and Zoun all the other players are mixed in place 2 to 4 as well and maybe Rogue, Cure and Zoun are not present at all or present in some. Is that the example? If so, there is definitely context missing to say who the greater player is; as well as alternative 3 and 4. Alternative 3 would be a player who does not win every tournament but 3 people constantly are in the top 4. Alternative 4 would be a player winning all tournaments, while place 2 to 4 is constantly changing. Can we deduce from the described alternatives 1 and 2 who is the better player? Only with context: - why are there always the same 4 on top in alternative 1? - Is their level higher, lower or equal than the player's level in alternative 2 overall (could be because of region locks)? - is the player pool equally big? - for the different metrics: tournament score will score alternative 1 higher or any other analysis that ratios 1st place better (like every tournament does according to prize money) Efficiency score as well, as it is depending on tournament score Match win rate... not necessarily. If in alternative 1 the players loses 2 games in a group stage and the alternative-2-player loses semis or finals tournament win %, yes of course, as alternative 2 won less for the same tournament count. Average placement as well. Now going from this hypothetical to my article: Every time contextual differences occurred, I tried to value them and adjust. Region locks were either completely left out or heavily penalized. Era was taken into account. Most of the time, when I had to make a decision (like including the inflated match win rates from Koreans versus Serral), I decided to lesser or more extent against Serral. Alternative 1 is on point (though i'd rather use Clem as example), alternative 2 is almost there but instead of having anyone else show up consistently everyone else either straight up does not repeat or at most appears 2 times out of 10 tournaments in top 4. To ground it in real albeit very mean example: would soO be GOAT if his Kong streak didn't phase him and he would go on to place second in every big tournament for a few more years until retirement/military service? The point here is that no matter what criteria you use, if alternative 1 is favoured, it is likely you are measuring dominance but not exactly greatness. I dunno what you mean by "kong streak phased him". soO actually went 4-0 in the finals of the four premier/major events he reached from 2019 onwards. He won IEM Katowice 2019, Master's Coliseum 4 and 5 and TSL 5 in 2020. soO's issue was that he was quite poor in the late game and, as SC2 aged and games began to reach said stage at an earlier and earlier point, his results began to suffer more and more. Also we should remember when soO had that GSL kong streak in HOTS zerg wasn't as strong as it is in LOTV. I think people underrestimate how great soO was in 2014. Way to make this about Zerg being OP in Lotv!
|
Speaking of reading comprehension, I read it as scenario 1: 1 player wins tournaments with the same players in top 4, which might suggest surperior players, but I would say a stagnant pool. VS. scenario 2: 1 player wins tournaments, but the top 4 has different players in it,meveryone except the player winning.
Hots zerg was indeed harder (game wasn't figured out so you had to scout for everything) and balanced (they didn't auto win late game).
I really don't mind a life&inno top 2. Current sc2 is about 5\20 as that guy said, which is a 4x multiplier for the hardest era.
I can appreciate the skill set achieved by playing the same game for a long time, but what this serves as is a barrier for new talents to come in a break the mold. So, you might say, then they don't deserve it, but the fact is the talent pool is so low today, and yet we see a guy like shin get top 4 and he might even win one day, but this guy would probably be c tier in peak sc2. Just look at a guy like dream, incredibly fast guy and hanged with the best, but he didn't rly achieve anything, were he to peak now instead he would be a menace for all of the top guys today to deal with. He might not last for a long time, but someone like him could completely disrupt the system, in my view.
For me what is most important is tournament score, but there is the caveat that by just hanging around this score increases and there isn't rly a fair way to deal with that. A guy like dark probably isn't a top 5 in terms of splash of what he did, he never really stood head and tails above the rest, but he sorta hung around getting top 5 in terms of achievements, and while I appreciate the consistency required there, I don't think he is a top 5 talented player or anything like that.
With your methodology and rogue being subpar, does that prove that a current player who is not a great can achieve amazing tournament score, and does that not prove then, using your methodology that this is a much weaker era?
Also, I think 3rd place should be almost the same as 2nd place, if we played double elimination, they would be same place, but since the tournament organizers award more money for 2nd, it should be slightly higher.
In my view tournaments shouldn't be looked at how stacked they were, but rather how stacked a pool this tournament represents.
So, the maru and inno WC wins, while they weren't stacked, there rly isn't any excuse that the best players didn't go, this was where the money was at and every region was represented by qualifiers, so as the title suggest it's a world championship. Still, a system that acrues points over a full year is more detailed, and I would award such a tournament higher in that regard. For, instance the entire tournament structure was surrounding blizzcon and so no tournament is more important.
|
Northern Ireland25613 Posts
On June 19 2025 16:57 ejozl wrote:
Speaking of reading comprehension, I read it as scenario 1: 1 player wins tournaments with the same players in top 4, which might suggest surperior players, but I would say a stagnant pool. VS. scenario 2: 1 player wins tournaments, but the top 4 has different players in it,meveryone except the player winning.
It depends who those players are to a pretty big degree.
If we look at tennis, and Federer/Nadal/Djokovic/Murray. Those cats are dominating for aeons, they’re very frequently the top 4 in the big tournaments. But it’s generally not considered a weak era, just players who are that level above, showing it. Arguably the 3 GOATs, at their peaks at roughly the same time, and Andy Murray, who still got slams, spent time as world number 1, 2 Olympic golds and is pretty high up there in non-slam tournament wins and win percentages all-time (11th for percentage, 14th for ATP titles, 5th for Masters 1000 titles in the Open Era). So even the weakest guy, with what he accomplished despite playing at the same time as arguably the top 3 GOATs, is still showing top 10/20 numbers, and almost certainly would have done better than that again in a different timeframe.
Whereas woman’s tennis in recent years, good luck predicting the top 4 of a tournament successfully. It’s almost too volatile in ways, it makes for exciting tournaments, but without consistently dominant players those storylines and rivalries don’t really develop. You haven’t really had that dominant player since Serena Williams was doing her thing.
Anyway that aside, if we actually look at tournaments over the last while, the Ro4 lineups aren’t as constant and predictable as I think perception would have one believe.
It’s certainly lacking quite the depth of other eras in terms of realistic tournament winners, which I think yeah, can’t argue with that. But it’s not nearly as shallow as some argue. Going back a few years admittedly we’ve had the likes of Trap, Zest, Zoun, Gumiho, Cure, Solar, Bunny, Byun (and more probably, going off memory) either winning or making the finals of offline events. Oliveira’s famous Katowice of course. Classic’s just won GSL this season. Dark’s usually floating around of course, and Reynor can win any tournament even if he’s not always making it deep. herO’s usually a presence at the business end of tournaments, albeit not always.
It’s really only Serral who’s almost certainly a lock for a Ro4 finish over the 2018 thru now span. Maru had pretty long spells where he was doing so, his monster 2018, he was incredibly strong in the Covid era and a bit beyond. Clem’s looking to be doing that in his current incarnation.
On June 19 2025 16:57 ejozl wrote:
Hots zerg was indeed harder (game wasn't figured out so you had to scout for everything) and balanced (they didn't auto win late game).
I really don't mind a life&inno top 2. Current sc2 is about 5\20 as that guy said, which is a 4x multiplier for the hardest era.
This goes both ways. I think there’s value in both being ahead of the curve in some way in a game that isn’t as much figured out, and staying at the top of a much more fleshed-out game where it’s harder to get an edge.
Balance? It fluctuates. Not always equally, sure. There were periods of imbalance in WoL and HoTS too, and that has an effect as well.
4x multiplier is crazy IMO.
I love Inno, speak not of Voldemort, but why wasn’t he smashing faces in the ‘weaker era’? I think he’s a case for #3, depending on my mood I stick him there, I just don’t see him having a case that beats Serral and Maru. Serral’s been (prolifically) winning tournaments every year for 7+ years now, Maru even longer again.
On June 19 2025 16:57 ejozl wrote:
With your methodology and rogue being subpar, does that prove that a current player who is not a great can achieve amazing tournament score, and does that not prove then, using your methodology that this is a much weaker era?
No, how does it show that? It shows that Rogue, despite his stellar results in terms of trophies, lags way behind some of the other GOAT contenders in terms of consistency.
If anything, it’s an argument in the opposite direction. If Rogue is good enough to win the big prizes on his day, but also exit tournaments earlier than expected, that indicates to me it’s still pretty tough at the top.
It can’t simultaneously be a weaker era because it’s the same guys in the Ro4 all the time, but also be a weaker era because Rogue won huge titles but wasn’t in the Ro4 every tournament.
On June 19 2025 16:57 ejozl wrote:
Also, I think 3rd place should be almost the same as 2nd place, if we played double elimination, they would be same place, but since the tournament organizers award more money for 2nd, it should be slightly higher.
In my view tournaments shouldn't be looked at how stacked they were, but rather how stacked a pool this tournament represents.
So, the maru and inno WC wins, while they weren't stacked, there rly isn't any excuse that the best players didn't go, this was where the money was at and every region was represented by qualifiers, so as the title suggest it's a world championship. Still, a system that acrues points over a full year is more detailed, and I would award such a tournament higher in that regard. For, instance the entire tournament structure was surrounding blizzcon and so no tournament is more important.
The entire crux of the ‘everything that isn’t Kespa era was a weaker era’ argument is that Proleague and Korean Individual Leagues were ruthlessly stacked with basically all of the best players, and more of them.
I’ve no particular issue with that line of reasoning, but people start to get inconsistent in application.
|
Imo Inno's impossible to argue over Maru simply because most of his achievements were also after the kespa players quit. They were earlier than Marus but still post Kespa. Even his 2016 IEM win was after Kespa quit SC2. Even with a 4x multiplier for Kespa era I don't see how he's above Maru unless you count his late 2016 and 2017 results as Kespa era which they simply aren't. Life I can see.
|
These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis.
Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this.
Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there.
That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is.
Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here.
|
Wombat might be right about the rogue fact, I'm unsure, anyways, I think he's a god. For me showing excellence despite lower average lvl is almost more impressive. I love stats for that as well, mb saying he's almost smurfing, because his ladder rank is that low, meanwhile he wins gsls.
|
On June 21 2025 01:40 TheDougler wrote:Show nested quote +These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis. Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this. Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there. That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is. Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here.
Claiming the conclusion holds up despite acknowledging the methodology is flawed is classic confirmation bias. If you're arguing the methodology is flawed, you can't on the same brush accept the conclusion, much less that the conclusion is somehow indicative of how "dominant Serral is."
|
Northern Ireland25613 Posts
On June 21 2025 04:12 lokol4890 wrote:Show nested quote +On June 21 2025 01:40 TheDougler wrote:These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis. Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this. Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there. That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is. Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here. Claiming the conclusion holds up despite acknowledging the methodology is flawed is classic confirmation bias. If you're arguing the methodology is flawed, you can't on the same brush accept the conclusion, much less that the conclusion is somehow indicative of how "dominant Serral is."
Premo has already shown some of Serral’s freaky numbers, and shown his working.
In terms of how to weight his categories for a final ranking, yeah that’s flawed. I think it’s flawed. Hell Premo said its flawed and he wasn’t sure how to weight it, hence why he got GPT to do it. Which he also conceded wasn’t ideal
But I think there’s more than enough of a case in the rest of the analysis that Serral is certainly SC2’s most dominant player relative to his contemporary competition.
Some will still feel that isn’t enough to make him the actual Greatest of All Time; which is fine by me!
I do however think it’s actually absurd to argue that he isnt SC2’s most dominant and consistent player. His numbers are simply too good
|
On June 21 2025 04:12 lokol4890 wrote:Show nested quote +On June 21 2025 01:40 TheDougler wrote:These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis. Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this. Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there. That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is. Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here. Claiming the conclusion holds up despite acknowledging the methodology is flawed is classic confirmation bias. If you're arguing the methodology is flawed, you can't on the same brush accept the conclusion, much less that the conclusion is somehow indicative of how "dominant Serral is."
Of course you can. If you ask 100 people without tastebuds if sugar tastes sweet and they say yes, you can absolutely demolish the metholodogy of this experiment, but you can't really argue that the conclusion is still correct.
|
WombaT's take is quite accurate, so not much to add. Like it was pointed out before: The weighting isn't on point... but each and every weighting that does not hypertune efficiency, will lead to the same conclusion, as Serral is so consistent in all metrics. That is the biggest take away of this evaluation, in my opinion.
While looking into ejozl's suggestion in regards to double-elim's 3rd place being awarded the same as 2nd place I discovered some connection errors in my excel list.
Maru and Rogue lost some points as I somehow had their 3rd/4th connected to 3rd place. Maru went from 64.14 to 63.75; Rogue from 30.70 to 30.55.
A quick look at the excel sheets suggests that INnoVation and Serral would benefit both the most when changing 3rd place closer to 2nd place (both have 5 each), but INnoVation would earn slightly more as one of his 3rds was in 2013 which has the era-multiplier on top. It is an interesting idea, but it won't change much in the big picture.
I think for the rework, I will mostly focus on the weighting. The consensus so far is that tournament score is the most important metric, followed by tournament win percentage. For the latter, there are discussions going on, if only prime years should be counted, which I would find unfair towards players who were able to keep up high efficiency for many years. I think looking at all years makes also a better balance, as "simply sticking around" is already valued in the tournament score, as ejozl wrote. Looking overall at careers, efficiency and duration both have their place, which makes for a more well rounded result.
Aligulac seems to strike a nerve with some people, but while it has its issues, I can totally see a value for tipping the scale if things are close between players at the end.
Perhaps something along those lines:
Aligulac Rank: 5%
Match Win Rate: 5%
Tournament win %: 36%
Average placement: 5%
Tournament Score: 44%
Efficiency Score: 5%
Like this, I keep the same ratio as Miz (0,55 to 0,45) and include the other valuable metrics as tipping points that also carry value.
I could also make different versions to see how changing weightings results in different outcomes. For example:
Version 1: As written above, only prime years for tournament win %
Version 2: As written above, all years for tournament win %
Version 3: Without Aligulac, only prime years for tournament win % (Tournament score 46,75% and Tournament win % 38,25%; the other three 5%)
Version 4: Without Aligulac, all years for tournament win % (Tournament score 46,75% and Tournament win % 38,25%; the other three 5%)
Any thoughts?
|
On June 21 2025 06:05 Balnazza wrote:Show nested quote +On June 21 2025 04:12 lokol4890 wrote:On June 21 2025 01:40 TheDougler wrote:These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis. Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this. Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there. That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is. Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here. Claiming the conclusion holds up despite acknowledging the methodology is flawed is classic confirmation bias. If you're arguing the methodology is flawed, you can't on the same brush accept the conclusion, much less that the conclusion is somehow indicative of how "dominant Serral is." Of course you can. If you ask 100 people without tastebuds if sugar tastes sweet and they say yes, you can absolutely demolish the metholodogy of this experiment, but you can't really argue that the conclusion is still correct.
In no professional field would anyone accept a conclusion of a statistical study with a flawed conclusion. That just doesn't happen, ever. But I guess things are different when the conclusion is serral = goat, got it.
|
Northern Ireland25613 Posts
On June 21 2025 09:22 lokol4890 wrote:Show nested quote +On June 21 2025 06:05 Balnazza wrote:On June 21 2025 04:12 lokol4890 wrote:On June 21 2025 01:40 TheDougler wrote:These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis. Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this. Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there. That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is. Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here. Claiming the conclusion holds up despite acknowledging the methodology is flawed is classic confirmation bias. If you're arguing the methodology is flawed, you can't on the same brush accept the conclusion, much less that the conclusion is somehow indicative of how "dominant Serral is." Of course you can. If you ask 100 people without tastebuds if sugar tastes sweet and they say yes, you can absolutely demolish the metholodogy of this experiment, but you can't really argue that the conclusion is still correct. In no professional field would anyone accept a conclusion of a statistical study with a flawed conclusion. That just doesn't happen, ever. But I guess things are different when the conclusion is serral = goat, got it.
Way to ignore the last few posts.
Serral’s got the best win percentages in terms of overall performance, outside of Life he’s got the best ratio of tournaments won to tournaments played. He’s got an average placement in tournaments of better than Ro4, over approaching 8 years. He’s won the most Premier tournaments, even excluding WCS region-locked ones.
All of which Premo laid out exhaustively.
I think his final analysis is flawed, as I said earlier, as indeed he conceded.
|
On June 21 2025 09:22 lokol4890 wrote:Show nested quote +On June 21 2025 06:05 Balnazza wrote:On June 21 2025 04:12 lokol4890 wrote:On June 21 2025 01:40 TheDougler wrote:These are the weightings CHAT GTP suggested:
Aligulac Rank: 20%
Match Win Rate: 15%
Tournament win %: 17,5%
Average placement: 15%
Tournament Score: 22,5%
Efficiency Score: 10%
Although I don’t fully agree with all of the suggested weights, I adopted them as-is - both for simplicity and because the final outcome would only shift significantly under an extreme weighting of efficiency.
So finally, here is the normalized and weighted final result of this analysis. Bro, why!? I agree with your conclusion, but "I used AI because it sounded cool" is just terrible methodology. You didn't make the case for why you needed ChatGPT for this, and so of course people are going to critique this. Also Aligulac being scored higher than tournament % and match win rate is dumb for multiple reasons. You're going to have massive colinearity between match win % and aligulac, in short you're double counting match win % there. That being said, I do think that the conclusion holds up despite the flawed methodology, which is evidence of just how dominant Serral is. Lastly, this was clearly a lot of work and I thank you for that. But, yeah you're double counting and the use of ChatGPT was pretty questionable here. Claiming the conclusion holds up despite acknowledging the methodology is flawed is classic confirmation bias. If you're arguing the methodology is flawed, you can't on the same brush accept the conclusion, much less that the conclusion is somehow indicative of how "dominant Serral is." Of course you can. If you ask 100 people without tastebuds if sugar tastes sweet and they say yes, you can absolutely demolish the metholodogy of this experiment, but you can't really argue that the conclusion is still correct. In no professional field would anyone accept a conclusion of a statistical study with a flawed conclusion. That just doesn't happen, ever. But I guess things are different when the conclusion is serral = goat, got it.
I thought I explained it enough, but I will try to word it differently.
A Greatest of all Time needs to have three key characteristics, when looking at statistics: Duration (a minimum threshold should be cleared), Efficiency and Dominance. I tried to think of how these characteristics could be measured the best and arrived at 6 metrics.
I asked several times - 1 year ago, when the first article came out - as well as this time, if anyone could think of another important metric, but it became clear that the tournament score was probably the most important to most people. A 7th one was never mentioned (at least I can't remember).
As I said: People had lots of issues with Aligulac because of cross-regional comparison problems, but from my understanding this shouldn't have been a problem in 2017, the only time frame when a non-foreigner became relevant in this metric. The time frame which we speak about is at most a couple of weeks and even then - from what I understand - did the machine not do anything wrong. Aligulac may not be perfect, but dismissing it entirely isn't the correct way forward, from my point of view.
Back to methodology: The 6 described metrics all have methodologies through which they are established. I tried to use contextualized reasoning like an era-multiplier and explained extensively when subjective gradings were necessary, for example in the grading of tournaments.
The results needed to be normalized... so far, the methodology is sound, although one can discuss if one or the other tournament could be counted more/less, if the era-multiplier is enough/too little, how the placements weigh against each other (like ejozl's suggestion) or if only prime years are counted.
Except for the prime year-idea or the era-multiplier, it is my sincere estimate, that nothing would change too much in the final score if we tweaked around with the numbers with logical intent.
I think that the methodology up to this point is extremely clear and reasonable.
Like JJH777 pointed out, one could discuss, if the pre-2018 era gets another subdivision of era-multipliers.
BUT: Even this would probably "only" lead to Maru come closer in the tournament score and him being a tad more efficient.
Now comes the part, which I conceded was handled badly: The weighting.
As I explained several times: This part adds more subjectivity and personal liking but the core take away here is: Yes, the weighting I published is off, but the final result can only change if we hypertune efficiency or the era-multiplier so that Life gets ahead of Serral. But then Maru and Rogue will lose out in the process too against Life. And as no one saw efficiency as a super interesting or important metric (neither here or on Reddit), the final result seemed pretty clear: As Serral is so good among all important metrics, statistically you can't talk him out of the top spot, as he is first among most analyzes. And if he rarely is not first, he is a close second, whereas others have sub-par results mixed in.
A thought I did not mention so far: In my opinion, Serral's lead will only grow if I add more players into the analysis.
Why is that: Well, I don't see someone taking Serral's top spot in any metric. Also Life's efficiency and Serral's 2nd place here is nothing that can be taken from my estimation. Now imagine if we include Dark, Reynor, soO, sOs, Zoun, Zest, herO, Jaedong, MarineLord, Clem and I don't know who else. I bet you that some of these are better than the players who ranked 3rd in some of these metrics (similar to Rogue's results in the first and 2nd article). That means that Serral's lead will naturally grow in a final result, as relatively the other already examined players got worse through the addition of new ones.
Anyway: If you try to put a super heavy emphasis on 2012-2015, then the claim of Rogue and Maru vanishes, if you max it out to a degree in which Life comes close or to or overtakes Serral in the entire weighting.
Another way would be to only look at prime years and leave Aligulac out entirely (Version 3 of my previous post). But even that wouldn't lead to Serral taking too much of a hit, as his 2014-2017 weren't the best esports years in SC2 history, to be quite frank (context: he wasn't a full time pro back then). So looking only at prime years will help him too. I think version 3 or 4 are the most-unfavorable-to-Serral-options but there are limits to which it is reasonably to tweak the numbers.
That being said: Yes, the final methodology was flawed as I conceded several times. But it is easy to see, how this flaw does not change the final verdict. But yeah, if you want to give me your personal weighting, lolol4890, I can easily calculate the outcome (although the normalized results are posted and you could do it yourself probably in less than 10 minutes as well).
Another note: Yes, more subjective traits like personality, impact, legacy or circumstances can play a role too. These things - in SCII terms - could be Maru's insanely long career, Life's match-fixing, Mvp dominating the early stages, Serral transforming the idea of the dominant Koreans. To me, they are nice stories, but they depend too much on personal preferences or influence. Statistics blend out personal biases and prohibit unintended, illogical application of standards, although weightings or multipliers can of course be discussed.
@WombaT: I think Maru now won 18 and Serral 17 when locked tournaments are excluded, although both were tied in Premier Tournament wins at the time of the article. But context here is important: Maru won 10 of these tournaments (all his GSLs, StarsWar 11, Africa GSL) without Serral as well as mostly Clem and Reynor participating.
|
|
|
|
|
|
EPT
