by Caspar “Mobiusman”  van der Sman (Twitter: @moby_osman)

(I had hoped to get this article published before Smash Summit 3, but unfortunately I wasn’t able to get it finished in the short window after Canada Cup. Please just pretend Summit singles hasn’t started yet, or feel free to laugh at any results that run completely counter to the stats!)

October has been by far the most volatile month since I started collecting results, largely due to the long string of unlikely results at The Big House 6. Before going any further, I think it’s important to consider one of the differences between a data-based and an opinion-based ranking. When a player like Plup has one extreme outlier performance – like losing to Zain – its impact on his perceived skill should be relatively minor, if his results are otherwise consistent. It shouldn’t be ignored, but it is easy to explain away if there are other, unquantifiable factors involved, like rust, illness, or even motivation. Glicko meanwhile makes no such considerations. I will remove results that are obviously caused by sandbagging, but a result that is simply caused by a particular player having a bad day can warp these rankings much more than they would SSBMRank. Whether you want to consider this volatility a bug or a feature is up to you, but please try to keep it in mind if you’re wondering why the changes this month are more dramatic than you expected.

For any new readers, this is a ranking using scores generated by the Glicko chess algorithm, with a data set consisting of large tournaments from the past 12 months. I went into a bit of detail about Glicko and why it fits for Melee in my first article, which you can check here. However, I also explained that these results should be taken with a pinch of salt and considered as just another perspective on the metagame – not an attempt at an authoritative, objective ranking. A couple of things to note on this list: Hax, aMSa, and KirbyKaze would likely be scored high enough to make the top 30, but they are excluded due to lack of data. In addition, a few results have been removed due to sandbagging (mainly Leffen). The “Previous” ranking refers to the September rankings which I posted on twitter here.

At the very top, Armada and Hungrybox trade places for the third consecutive month. This time Armada’s lead is a little bigger, due to Hbox dropping another set to someone outside of the top 3 (Mew2King). Mang0 still hasn’t closed the gap, but his recent winning record over Armada has done a lot to reduce it, and it’s easy to imagine him overtaking one or even both players after Smash Summit. The second half of the top 10 sees a lot of movement this month, with traditional favourites like Plup, Shroomed, Westballz and Axe all losing ground to the three players with a huge amount of momentum this year – SFAT, Swedish, and Wizzrobe. While the same cannot be said of Westballz, the other three players I mentioned have for a long time been extremely consistent performers whose results can be predicted with reasonable confidence. But TBH6 changed all that, with Shroomed losing two Sheik dittos to KirbyKaze and Swedish, Plup losing to Shroomed and Zain, and Axe losing to R2DLiu and Zhu (adding to his other outlier loss to Mafia earlier in the year). It’s hard to overstate how uncharacteristic these results are. Of course it’s entirely possible that next month we will be back to the old order of things, but for now Swedish and Wizzrobe’s consistent results and SFAT’s multiple wins over Hbox put them on top.

A little further down, the world’s favourite non-Summit-attendee Ice explodes up the list after beating SFAT and Silent Wolf at TBH6, and following it up by double eliminating Leffen at Eclipse 2 – perhaps making him the new Fox Ditto Master (TM). Most people expect Leffen to improve from this performance as he dusts off his long-dormant fox, but this is still a very significant result and an indication that even in Europe, the gap is slowly shrinking. PewPewU meanwhile is continuing to ride his momentum from Evo, double eliminating Westballz at Blood for Blood and taking the whole tournament after trading sets with Nintendude. Someone on Reddit can correct me if I’m wrong, but I believe this makes B4B the largest tournament ever won by a solo Marth. This time last year, some people were sceptical of Marth’s role in the metagame, but PPU and The Moon are showing that the sword is just as sharp now as it was in 2007.

So, let’s talk about Laudandus (I know he’s been waiting for me to write that sentence). It is counterintuitive for him to jump up the rankings this month when his best performance (at Pound 6) was back in April, and he hasn’t had any recent results that are on the same scale – his best wins of the last 6 months being against n0ne, Cactuar, and Gravy. The simple answer is that Laudandus has been more consistent than his rivals, with only one “bad” loss since Genesis 3 (to Infinite Numbers). The more complex answer is that his lack of losses is a direct result of his comparative lack of data (note his deviation is almost as high as Leffen’s). Laudandus started off the year by beating Swedish, Westballz, Nintendude, and S2J, and while many people might consider those results atypical, there is not much in the data set to indicate that that is the case. I don’t want to give the impression that I’m writing this article to explain why Laudandus’ rating is wrong, since I don’t think it is. But again, it’s interesting to consider the way different people can be rewarded by different systems.

The rest of the top 25 remains largely the same with a few places shuffled around. Silent Wolf gets a big boost from his win over Mew2King, while (as I predicted previously) The Moon is looking very strong, double eliminating Lucky from B4B. While Lucky himself has only lost to good players, he has also not picked up any wins in the top 30 in the last 3 months (within the data set). His SoCal compatriots MacD and S2J also slide a few places along with him, but neither of them have had bad enough Octobers for it to mean too much – the gap between 20 and 25 is even smaller than usual this month.

dizzkidboogie is long overdue a place on this list, having been bubbling under around the top 35 area for several months. It’s worth noting that his absurd performance at Olympus, where he beat Prof, Lucky, and Moon twice, was not even included the data set. Even without those results, dizz still has recent wins over Silent Wolf and MacD, and has actually displaced the ever-unpredictable Wobbles in the Ice Climbers hierarchy. Bladewise has had a much quieter season, with dizz, Zhu, and Mike Haze (this month’s #31) the biggest names on his hitlist. But looking back over his results, they have been extremely consistent, with only two losses outside of the top 30 since Genesis (CDK and DJ Nintendo). That means he has a good resumé of wins over strong players who don’t quite make the cut, including Milkman, Trulliam, lloD, Azusa, and HugS. It’s not clear whether Bladewise can convert this consistency into causing a real upset, but for now he acts as a gatekeeper into the top tiers.

November will bring two of the most significant events of the year, particularly as we approach the SSBMRank voting season. All eyes will be on Leffen for Smash Summit 3, when we will finally find out whether he can repeat his performance from GOML. But with the top 10 in general taking such a beating at TBH6, Summit will also be important in establishing the new hierarchy going forward. DreamHack Winter last year gave Europe the chance it had been craving to go head-to-head with a range of US players. It didn’t really go too well – Tekk beat Duck and Android beat S2J, but Leffen’s infamous 9th place finish meant there were a total of 7 Americans in the top 8. Europe has grown a lot since then, with both n0ne and Chillindude failing to make top 8 at the European majors they flew to in 2016. Hopefully this year’s DreamHack will be as international as the last – as Melee continues to grow, opportunities to have different scenes meet become more and more important.