You should first of all take a look at the Top16, as it will give you better background knowledge of how to interpret the following:
Expected top cut value: 16 / 115 = 13,9%
Any Pokemon that has a value higher than 13,9% increased the chance of cutting, if you had it. For example, Kangaskhan has a value of 5 out of 46 made top cut. That equals 10,9%, which is lower than 13,9%. Therefore, having a Kangaskhan in the team decreased the chance of top cutting.
For example: Garchomp was used 67 times, which means it was on 58,3 % of the teams. 14,9 % of the Garchomp were in Top16, which means that 62,5 % of the teams in the Top16 had a Garchomp.
Raw Usage - Pokemon - % of Pokémon that top cut - % in teams of top cut - % in all teams
67 Garchomp - 14,9 % Cut - in 62,5
% of the teams in Cut - in 58,3 % of the teams
61 Rotom-W - 13,1 % Cut - in 50 % of the teams in Cut - in 53,0 %
of the teams
Less than 50% of the teams had these Pokemon
Less than 50% of the teams had these Pokemon
52 Tyranitar - 11,5 % Cut - in 37,5 % of the teams in Cut - in 45,2
% of the teams
47 Salamence - 17 % Cut - in 50 % of the teams in Cut - in 40,9 % of the
teams
46 Kangaskhan - 10,9 % Cut - in 31,3 % of the teams in Cut - in 40 % of
the teams
45 Aegislash - 17,8 % Cut - in 50 % of the teams in Cut - in 39,1 % of
the teams
42 Talonflame - 14,3 % Cut - in 37,5 % of the teams in Cut - in 36,5 %
of the teams
Less than 25% of the teams had these
Pokemon
28 Gardevoir - 17,9 % Cut - in 31,3 % of the teams in Cut - in 24,3 % of
the teams
18 Mawile - 33,3 % Cut - in 37,5 % of the teams in Cut - in 15,7 % of
the teams
18 Ferrothorn - 0 % Cut - in 0 % of the teams in Cut - in 15,7 % of the
teams
17 Charizard - 11,8 % Cut - in 12,5 % of the teams in Cut - in 14,8 % of
the teams
15 Meowstic - 13,3 % Cut - in 12,5 % of the teams in Cut - in 13,0 % of
the teams
14 Amoonguss - 7,1 % Cut - in 6,25 % of the teams in Cut - in 12,2 % of
the teams
13 Manectric - 0 % Cut - in 0 % of the teams in Cut - in 11,3 % of the
teams
12 Gengar - 8,3 % Cut - in 6,25 % of the teams in Cut - in 10,4 % of the
teams
12 Politoed - 16,7 % Cut - in 12,5 % of the teams in Cut - in 10,4 % of
the teams
Less than 10% of the teams had these Pokemon
11 Scrafty - 18,2 % Cut - in 12,5 % of the teams in Cut - in 9,6 % of
the teams
11 Kingdra - 18,2 % Cut - in 12,5 % of the teams in Cut - in 9,6 % of
the teams
10 Gyarados - 20 % Cut - in 12,5 % of the teams in Cut - in 8,7 % of the
teams
9 Azumarill - 0 % Cut - in 0 % of the teams in Cut - in 7,8 % of the
teams
9 Mienshao - 0 % Cut - in 0 % of the teams in Cut - in 7,8 % of the
teams
8 Abomasnow - 25 % Cut - in 12,5 % of the teams in Cut - in % of the
teams
7 Rotom-H - 0% Cut - in 0 % of the teams in Cut - in 7,0 % of the teams
6 Venusaur - 33,3 % Cut - in 12,5 % of the teams in Cut - in 5,2 % of
the teams
5 Rhyperior 40 % Cut - in 12,5 % of the teams in Cut - in 4,3 % of the
teams
5 Klefki 0 % Cut - in 0 % of the teams in Cut - in 4,3 % of the teams
4 Scizor, Bisharp, Hydreigon, Sableye, Aromatisse, Malamar - in 3,5 % of
the teams
3 Rotom-C, Gothitelle, Conkeldurr, Weavile, Torkoal, Reuniclus - in 2,6
% of the teams
Wow! Such numbers.
Rotom-W on the other hand was not as present in the Top16 as one would expect from it's raw number amongst all participants. 8 Rotom-W in Top16 is still as much as expected and not significantly lower.
Trends are seen with Pokémon such as Salamence, Aegislash, Gardevoir and Mawile. All of them had a higher appearance in Top16 than within the 115 participants. It looks like all of these Pokémon worked very well in the metagame the Japanese had and all of them had higher numbers in Top16 than expected. Especially Mawile stands out, as 6 out of only 18 Mawile in the whole tournament, managed to get out of their groups and reach Top16 - a phenomenal amount.
On the other hand, there is Pokémon like Kangaskhan, Amoonguss and especially Ferrothorn, Manectric, Mienshao and Azumarill. All of them were seen less Top16 than one would expect from their overall usage.
Particularly Ferrothorn and Manectric stand out, as neither of them managed to advance, even though their usage was pretty high with 15,7 % and 11,3 % respectively. It means that the Pokémon listed above did not work well in that specific tournament.
At least we found the Pokémon we felt like they were missing in the Top16 stats!
Of course not! When interpreting those stats, you have to keep in mind, that there is a lot more factors one has to consider. First of all, there is playing and luck obviously, but in addition to that, there is pairings and stuff like that. Therefore, it is very difficult to construe the data we've got, but certain tendencies can be seen and figured out!
I hope you can use this data to adapt a little bit or to find new threats that are good versus a majority of the Pokémon listed above!
Wow! Such numbers.
Yes, it is a little bit difficult to interpret those stats, because I was a little too eager and probably added too much content. However, each of those numbers has useful information, if you know how to understand them!
First of all, there have been 115 participants of whom we know the teams, but there were 123 participants in total. I don't know what happened to the 8 that are missing here, sorry. What I do know, though, is that over half of the teams had Garchomp and/or Rotom-W. Half of the teams! That is just insane and reflects very well how the big six are dominating the format at the moment.
Now onto deeper interpretation. Let's start with Garchomp again! While 58,3 % usage is already very high, this number even increased in the Top16, which means that the great number of Garchomp in Top16 is not only caused by the great amount of Garchomp in total! In fact, since the percentage in Top16 is higher, having a Garchomp increased your chances to get to Top16. Well, it's hard to put it that general, but that's what the number say.
Rotom-W on the other hand was not as present in the Top16 as one would expect from it's raw number amongst all participants. 8 Rotom-W in Top16 is still as much as expected and not significantly lower.
Trends are seen with Pokémon such as Salamence, Aegislash, Gardevoir and Mawile. All of them had a higher appearance in Top16 than within the 115 participants. It looks like all of these Pokémon worked very well in the metagame the Japanese had and all of them had higher numbers in Top16 than expected. Especially Mawile stands out, as 6 out of only 18 Mawile in the whole tournament, managed to get out of their groups and reach Top16 - a phenomenal amount.
On the other hand, there is Pokémon like Kangaskhan, Amoonguss and especially Ferrothorn, Manectric, Mienshao and Azumarill. All of them were seen less Top16 than one would expect from their overall usage.
Particularly Ferrothorn and Manectric stand out, as neither of them managed to advance, even though their usage was pretty high with 15,7 % and 11,3 % respectively. It means that the Pokémon listed above did not work well in that specific tournament.
At least we found the Pokémon we felt like they were missing in the Top16 stats!
Are these Pokémon lost then??
Of course not! When interpreting those stats, you have to keep in mind, that there is a lot more factors one has to consider. First of all, there is playing and luck obviously, but in addition to that, there is pairings and stuff like that. Therefore, it is very difficult to construe the data we've got, but certain tendencies can be seen and figured out!
I hope you can use this data to adapt a little bit or to find new threats that are good versus a majority of the Pokémon listed above!
- Markus
How come you didn't even mention Venusaur or Rhyperior in your analysis of the statistics? Surely you must think those numbers are relevant! A full ONE-THIRD of Venusaurs made it to top cut despite only being in 5.2% of teams! Clearly, the average tourney-goer is significantly undervaluing Venusaur.
ReplyDeleteYes, that is indeed true, but you have to keep in mind, that there were only 6 Venusaur and 5 Rhyperior in total. Imagine a tournament in which a certain group of the best players in the world built a team around a certain Pokemon and it would do very good in a tournament, statistically. However, this could also just be because the players were so good and not because the pokemon was superb. I left out parts with very low number evidence, as the fact that one third of venusaurs made top16 could also be random. The smaller the control sample, the less reliable a prediction is.
ReplyDeleteAlso, the 5,2% and the 33,3% do not have anything to do with each other. you can only compare 12,5% and 5,2%. The 33,3% refers to only the Venusaurs, while 5,2% refers to all pokemon used!
Thanks for the stats Markus!! :D
ReplyDelete