singstall
Smash Rookie
- Joined
- Apr 15, 2018
- Messages
- 1
- Location
- Woodbury, MN
- 3DS FC
- 3523-2496-2466
- Switch FC
- SW-4001-3948-9220
In case you're here for the graph, here you go: https://imgur.com/a/3NKwa
So for a while, I've had a nagging thought at the back of my brain that maybe recency bias gave Hungrybox the title of #1 in 2017. Sure, he was undoubtedly the best player in the world when it ended, but Armada felt so unbeatable during the first while that I wondered if Armada actually had the better resumé. After all, he won the two most prestigious tournaments of the year.
I can't really address prestige head on, I'm not the PGR, but I did come up with a two step way of addressing things.
First, I introduced dynasty points, a concept from FiveThirtyEight (although apparently one of their writers did use it in an NBA Insider article first) that acknowledges that the difference between getting first and getting third is a heck of a lot more meaningful than, say, the difference between getting third and fifth. Average placing is a pretty flawed statistic and I have a lot more to say about it, but let me lay out the approach in simple terms.
For each tournament placing you have, you get 1/x points, where x is the number of people who placed equal or better to you. This is pretty easy for two players who never placed lower than fourth all year, so x is just their placing. You get a whole point if you win! Some might say this is even a bit lenient, that placing fourth four times is not as good as winning once. Which, yeah, but it's still pretty nifty.
So which tournaments do I want to look at? This is going to be controversial, so I had to make hard choices that I'm sure not everyone will agree with. I only counted tournaments where at least two top six members were present and /playing seriously/. I don't count Mango at Frame Perfect Series 2, Leffen at Full Bloom 3 (tough call), or Plup playing secondaries (CEO Dreamland, Bad Moon Rising 2, even Shine 2017).
Also controversially, I decided to not give Hungrybox his locals, where he won quite a few CFL Smackdown titles over Plup. This also got rid of Smash Conference 69. Speaking of Smash Conference 69, I /have/ left in Genesis 4 for Hungrybox, despite him playing with a broken finger.
So what I did was I took these tournaments and calculated the average number of dynasty points per tournament. Remember that .5 is second place and 1 is first place.
Armada: (1+1+1+1/2+1/3+1+1+1/2+1/4+1/3+1/2+1)/12=.7014
Hungrybox: (1/2+1/4+1/2+1+1+1/3+1+1/3+1+1/2+1+1/3+1/2+1/2+1/4+1+1+1+1+1+1)/21=.7143
Then I wondered if maybe these got a little warped by the smaller events that Hungrybox attended, like Saints Gaming Live, Frame Perfect Series 2, Too Hot To Handle, etc (although Armada also has two wins with only Leffen present).
So I decided to try this with weighted averages. I decided to weight each tournament by the number of OTHER top six members present who were playing seriously.
Armada: (5+1+5+3/2+4/3+1+5+4/2+5/4+2/3+5/2+1)/41=.6646
Hungrybox: (1/2+5/4+5/2+1+2+1/3+4+3/3+4+1/2+1+5/3+2/2+1/2+3/4+3+4+5+2+1+5)/59=.7119
This is the data in the graph I made, which is again visualized here: https://imgur.com/a/3NKwa
Conclusion: Hungrybox actually does have a really solid claim to having the better 2017 resumé to the point of maybe having locked it down, although the gap between Armada and Hungrybox after, say, Super Smash Con 2017 got to look impossible to bridge.
The data fits the narrative almost hilariously well, with Summit 5 having been pretty much a must-win for Armada, exactly as he said at the time.
I appreciate anyone calling out if I messed up the data, because I just did this while watching the NBA Playoffs.
UPDATE: I made a graph of the entire top six during 2017, here:
So for a while, I've had a nagging thought at the back of my brain that maybe recency bias gave Hungrybox the title of #1 in 2017. Sure, he was undoubtedly the best player in the world when it ended, but Armada felt so unbeatable during the first while that I wondered if Armada actually had the better resumé. After all, he won the two most prestigious tournaments of the year.
I can't really address prestige head on, I'm not the PGR, but I did come up with a two step way of addressing things.
First, I introduced dynasty points, a concept from FiveThirtyEight (although apparently one of their writers did use it in an NBA Insider article first) that acknowledges that the difference between getting first and getting third is a heck of a lot more meaningful than, say, the difference between getting third and fifth. Average placing is a pretty flawed statistic and I have a lot more to say about it, but let me lay out the approach in simple terms.
For each tournament placing you have, you get 1/x points, where x is the number of people who placed equal or better to you. This is pretty easy for two players who never placed lower than fourth all year, so x is just their placing. You get a whole point if you win! Some might say this is even a bit lenient, that placing fourth four times is not as good as winning once. Which, yeah, but it's still pretty nifty.
So which tournaments do I want to look at? This is going to be controversial, so I had to make hard choices that I'm sure not everyone will agree with. I only counted tournaments where at least two top six members were present and /playing seriously/. I don't count Mango at Frame Perfect Series 2, Leffen at Full Bloom 3 (tough call), or Plup playing secondaries (CEO Dreamland, Bad Moon Rising 2, even Shine 2017).
Also controversially, I decided to not give Hungrybox his locals, where he won quite a few CFL Smackdown titles over Plup. This also got rid of Smash Conference 69. Speaking of Smash Conference 69, I /have/ left in Genesis 4 for Hungrybox, despite him playing with a broken finger.
So what I did was I took these tournaments and calculated the average number of dynasty points per tournament. Remember that .5 is second place and 1 is first place.
Armada: (1+1+1+1/2+1/3+1+1+1/2+1/4+1/3+1/2+1)/12=.7014
Hungrybox: (1/2+1/4+1/2+1+1+1/3+1+1/3+1+1/2+1+1/3+1/2+1/2+1/4+1+1+1+1+1+1)/21=.7143
Then I wondered if maybe these got a little warped by the smaller events that Hungrybox attended, like Saints Gaming Live, Frame Perfect Series 2, Too Hot To Handle, etc (although Armada also has two wins with only Leffen present).
So I decided to try this with weighted averages. I decided to weight each tournament by the number of OTHER top six members present who were playing seriously.
Armada: (5+1+5+3/2+4/3+1+5+4/2+5/4+2/3+5/2+1)/41=.6646
Hungrybox: (1/2+5/4+5/2+1+2+1/3+4+3/3+4+1/2+1+5/3+2/2+1/2+3/4+3+4+5+2+1+5)/59=.7119
This is the data in the graph I made, which is again visualized here: https://imgur.com/a/3NKwa
Conclusion: Hungrybox actually does have a really solid claim to having the better 2017 resumé to the point of maybe having locked it down, although the gap between Armada and Hungrybox after, say, Super Smash Con 2017 got to look impossible to bridge.
The data fits the narrative almost hilariously well, with Summit 5 having been pretty much a must-win for Armada, exactly as he said at the time.
I appreciate anyone calling out if I messed up the data, because I just did this while watching the NBA Playoffs.
UPDATE: I made a graph of the entire top six during 2017, here:
Last edited: