Welcome to the PGstats open beta!

We'd like to use cookies to help analyze and create a better experience for you.

You can find out more in our Privacy Policy and Terms of Service.

Which seeds make top 8s?

By Jack "Jackie Peanuts" Moore | 05/26/21

Last week, I took a look at a number of major tournaments from Melee, Ultimate, and Ultimate WiFi to analyze the consistency (or lack thereof) in these competitions. Unsurprisingly, that analysis, which you can read here, revealed that Ultimate's WiFi is significantly less consistent when it comes to seed performance than both Ultimate and Melee.

When looking at LAN only, Melee was slightly more consistent than Ultimate, but to a smaller degree than I expected given the intense skill gaps at top level Melee and the many years the competitive scene has had to establish itself. To further this analysis, I wanted to take another look at things from a slightly different angle, this time asking the question: Which seeds make top 8s?
click to expand
This chart looks at how many players of at least a certain seed attained top 8s in Melee, Ultimate, and the Ultimate WiFi qualifiers. Top 4 seeds made up 42% of Melee Top 8s (out of a possible 50%), with Top 4 seeds making up 34% of Ultimate WiFi top 8s and only 31% of Ultimate LAN top 8s. The gap between Melee and Ultimate maintains itself at every seed point. Top 8 seeds make up 70% of Melee top 8s but just 60% of Ultimate LAN top 8s and only 55% of Ultimate WiFi top 8s.
Interestingly, Top 4 seeds at Ultimate WiFi qualifiers actually performed slightly better than Top 4 seeds have at Ultimate S-Tiers, but that corrects itself pretty quickly, as seeds 5-8 performed much better offline than online.
The highest seed to ever make a Melee Top 8 was , the 35th seed who placed 5th at The Big House 8. At Ultimate S-Tiers, 7% of players in top 8s have been seeded 35th or worse, with , the 97th seed who placed 7th at Smash 'N' Splash 5, coming in as the lowest. For the Ultimate WiFi qualifiers, 9% of players in top 8s have been seeded 35th or worse, with , the 116th seed who placed 4th at the EU Qualifier, coming in as the lowest.
What this suggests to me is that while Melee may have a decent amount of inconsistency among its high-level players, its top tier is extremely consistent, exactly as you would expect from a game that has evolved into a competition between The Gods and the rest.
Meanwhile, Ultimate LAN's inconsistency stems in part from the fact that there hasn't been a group of top players that has broken away from the crowd enough to create consistent top 8s. I think this is reflected in the fact that most of the 1-8 seeds who aren't making it to Ultimate S-Tier top 8s are being replaced by seeds 9-13—the closest the Melee and Ultimate lines come to touching is at point 13: 81% of Melee top 8s have been made up of top 13 seeds, whereas the same is true of 78% of Ultimate top 8s.
We don't see this same replacement happen online. Far more players in the 16-32 range make top 8 at these tournaments, and marginally more players in the 64+ range make it as well. This could be due to the inconsistent nature of online play, and could also be due to the different populations attending online tournaments compared to offline tournaments as well as the region restrictions limiting top players from attending multiple events.
For the latest Smash analysis, follow @PGstats on Twitter.