EPTIn my mind, there are six, and that is all that matters. Because my opinions about a title based on opinion are actually facts. Even ask me.
Statistical Bonjwa Theory: the History of BW - Page 4
| Forum Index > BW General |
|
U_G_L_Y
United States516 Posts
In my mind, there are six, and that is all that matters. Because my opinions about a title based on opinion are actually facts. Even ask me. | ||
|
Release
United States4397 Posts
LeeSsang | ||
|
Xiphos
Canada7507 Posts
On August 01 2012 09:12 U_G_L_Y wrote: Based on this data, if you accept that Boxer was a Bonjwa, then you must also accept Jaedong. In my mind, there are six, and that is all that matters. Because my opinions about a title based on opinion are actually facts. Even ask me. Bisu was somewhat of a Bonjwa too. He defeated the previous one and he revolutionized PvZ not only once but twice. | ||
]343[
United States10328 Posts
On August 01 2012 09:30 Xiphos wrote: Bisu was somewhat of a Bonjwa too. He defeated the previous one and he revolutionized PvZ not only once but twice. unfortunately Mind ruined everything | ||
|
Forsy
Canada36 Posts
Bisu did dominate for sometime, and could be in contention, though many put him in the just under bonjwa tier. Stork, there will probably be more votes for him being closer to just under bonjwa tier altogether, but still possible. Fantasy, is possible, but same issues with Stork. Remember that the 5 mentioned are playing in an era beset with dominators, and it is possible that the era reached the limits of gameplay/brood war understanding (apart from slight mechanical improvement). Clearly, the first two should be considered Bonjwas, and the other 3, at least under review/contention. | ||
|
krndandaman
Mozambique16569 Posts
| ||
|
endy
Switzerland8970 Posts
On August 01 2012 11:00 krndandaman wrote: to be fair Flash's play had a certain artistry to it as well. the way he perfected simcity... flash grasped that concept better than anyone else in the history of broodwar. seeing the simcity and how every single building placement was preplanned gave me chills and an appreciation for flash's preparation and ingenious thinking. also, flash's starsense.... that thing gave me chills. particularly the one with going blind bunks against calm's unscouted 2hatch lurkling all-in. His starsense during Game 1 and game 4 of Korean Air S2 finals... amazing And his turret placement > all. Basically everyone would watch Flash play on a map and then copy his turrets placement. | ||
|
Demonhunter04
1530 Posts
On August 01 2012 00:23 hacklebeast wrote: Statistical Bonjwa Theory is a sick band name I second this. | ||
|
Crisium
United States1618 Posts
 Statistics are fun, but there's the subjective value of the games. Which is why Power Rank always held such a deep spot in my heart. Flash: Back-to-back-to-back dual finals in 2010, winning 4 of the 6. Such dominance was never achieved before, nor after. He also dominated SPL in 2010. And even as recently as late 2011 and early 2012, Flash was virtually unstoppable in SPL. It's the general talk of things that truly define a bonjwa. I remember that even after he made his third and fourth consecutive finals (losing to Effort, and defeating Jaedong) Koreans still denied him Bonjwa status. Several months later, Flash defeats Jaedong twice and they proclaim him God. How could they not? Flash and Jaedong were both far better than anyone else in the scene during this time. And Jaedong was helpless before Flash in the end. God > Jaedong > all. At this time, Bisu was considered slumping and not a top toss, and even Stork was only the #3 Protoss behind Kal and Free. None of these Protoss did anything notable in 2010. Effort couldn't follow up his success with another individual league run ever again. Flash and Jaedong were the only real men, and even Jaedong looked like crap when he got 3-0'd by a 14ccing Flash. TTT just can't express this dominance. Jaedong may have been second to Flash statistically by only a bit, but in late summer 2010 Flash was subjectively so much better it was scary. | ||
|
Yoshi Kirishima
United States10366 Posts
After knowing so much about BW history... it makes me want to cry that BW is ending. I'm a SC2 gamer but SC2 has its own flaws and BW is also a great game. Knowing all these players' stories, victories and defeats, ups and downs, and all that tragic stuff... is really beautiful. It's like people who may study old history of a country and look at the kings/emperors and etc. I don't think it's just me, but just thinking of the shifts in power between players (and the stories behind them) is so awesome. When I look at such graphs, and I look at 2000... it is the beginning. Then 2001, then 2002, you see shifts and shifts. Suddenly we're at 2007 and then 2012, and all's different. Progamers rise and fall. If these graphs are to represent skill, then it looks like Jaedong is the 5th bonjwa and Flash is the 6th. But then again, TBLS did really well during Jaedong's time, even if Jaedong's skill is higher than theirs for about a year. I guess it wasn't high enough (at least according to these graphs), as Flash becomes way higher than the rest for about 2 years. Surprised to see jangbi so low. I don't quite understand how the skill level is calculated though, but I would have thought jangbi is higher since he's been in 2 finals and beat flash. On a final note, since I've been following sc2 the whole time, i realize I still miss a lot. I don't know about the players that aren't at the top. Players like midas and sea, I don't know much about even though they are great. In SC2, it's only been about 2 years but I know the winners of all the GSLs, who is dominant and such, by the month. Looking at those 12 years of BW, it awes me how much more there is I don't know. I only know the general trends, by year. I don't know all the interesting stories from each month. Esports forever <3 | ||
|
Crisium
United States1618 Posts
| ||
|
Yoshi Kirishima
United States10366 Posts
Boxer was the first best player (aside from Grrrr, don't think he dominated long enough). Nada overcame Boxer, but Boxer's son iloveoov, who was older than Nada, overcame Nada later on. Then savior and flash/TBLS came in. But now boxer's grandson fantasy is here, and is #1 (well hopefully he beats jangbi this time lol). | ||
|
endy
Switzerland8970 Posts
On August 01 2012 13:03 Yoshi Kirishima wrote: Btw if Fantasy wins, I will be glad. Boxer was the first best player (aside from Grrrr, don't think he dominated long enough). Nada overcame Boxer, but Boxer's son iloveoov, who was older than Nada, overcame Nada later on. Then savior and flash/TBLS came in. But now boxer's grandson fantasy is here, and is #1 (well hopefully he beats jangbi this time lol). That reminds me how sad it is that Midas never won a Starleague  Especially when you take a look at the graph during Savior's domination era, Midas is pretty much the only one not too far behind Savior. | ||
|
ppshchik
United States862 Posts
| ||
|
Crisium
United States1618 Posts
On August 01 2012 12:46 Yoshi Kirishima wrote: Looking at those 12 years of BW, it awes me how much more there is I don't know. I only know the general trends, by year. I don't know all the interesting stories from each month. Check out the PRs! http://www.teamliquid.net/powerrank/ Scroll to the bottom, and read up. They begin at the height of Savior's dominance over the scene. Sure they are subjective, but rarely is there a truly unpopular #1 choice and only occasionally do some solid performers get snubbed from a low spot. | ||
|
gondolin
France332 Posts
On July 31 2012 15:19 d_ijk_stra wrote: In games like Chess and Go, which two players challenge against each other to determine who is the winner, probably the most famous rating system is the Elo rating system which is named after the Physicist Arpad Elo. This Elo system is actually also used in TLPD to rank both BW and SC2 gamers. But the system I used is the extension of Elo, called TrueSkill Through Time (TTT). Compared to the Elo system, this TTT system models how the skill of a player changes over time more explicitly. Actually, it was proposed by researchers in Microsoft to investigate the history of Chess; and as a SC nerd, I thought it would be fun to apply it to BW matchup records. I used this TTT system with almost no modification, so you may refer to the linked paper about assumptions I had to make for the analysis. I have to admit that the model relies on heavy assumptions. (So again there will be people claiming 'Statisticians are liars!') But using a statistical model such as Elo or TTT is often more useful than just plotting descriptive statistics such as win-rate each month, as the statistical model fully utilizes the information in our data while descriptive statistics may exaggerate certain aspect of data, possibly on purpose. At least in my opinion this TTT model is the standard way a Statistician may formulate the problem. Awesome analysis and graph!. I had always wanted to do a similar thing, but using Whole History Rating rather than TTT, I am so glad someone else did that instead of me! Since I am not a statistician, I was wondering: what is the difference between the two? They seem to use a very similar model, except that WHR use Newton approximations to converge quickly, so I guess it can afford to have a smaller unit interval. By the way what time interval did you use for the TTT? And what variability? | ||
|
d_ijk_stra
United States36 Posts
On August 01 2012 14:51 gondolin wrote: Awesome analysis and graph!. I had always wanted to do a similar thing, but using Whole History Rating rather than TTT, I am so glad someone else did that instead of me! Since I am not a statistician, I was wondering: what is the difference between the two? They seem to use a very similar model, except that WHR use Newton approximations to converge quickly, so I guess it can afford to have a smaller unit interval. By the way what time interval did you use for the TTT? And what variability? I am not very familiar with the WHR, but as a Statistician I would personally prefer TTT over WHR. Basically all these models are attempting to solve a special case of Bayesian classification, and it is known that for this class of problems MAP (Maximum A Posteriori) estimate which is used by WHR is not a good representation of posterior distribution than EP (Expectation Propagation) is. In terms of classification accuracy usually there is no big difference, and maybe that's why the WHR authors could've improved it in their paper, but here we are much more interested in the interpretation (who is a better player than the other) and thus supposedly TTT with EP should be more favorable. See this paper for reference. However, I could be biased because the authors of TTT paper are very accomplished researchers of the field and I love their work. Again, I do not have deep understanding of WHR and it is possible that WHR is a superior model at least in some context. I used each month as the time interval, but I could've implemented it without using time intervals and represent the date exactly. It is just more convenient to implement that way and I am a lazy fellow... :-) | ||
|
ore0z
Romania161 Posts
On August 01 2012 02:03 Fionn wrote: I thought Flash was proclaimed a bonjwa by the Korean press after he won the MSL/OSL in a span of a week? Pretty sure he's the 5th. http://www.teamliquid.net/forum/viewmessage.php?topic_id=152052 With this victory, you’ve risen to Bonjwa status. A Receiving a title like that is quite an honor but compared to the players who rose to that level before, I think I’m still lacking. Though it’s nice if fans call me that, I still think “Ultimate Weapon” is better and more appropriate for me right now. Edit: This quote is from the Korean Air Season 2 OSL, where he defeated Jaedong a second time. | ||
|
FlaShFTW
United States10310 Posts
someone said something on the lines on "bonjwa is a stupid word given to savior at his peak and lost meaning after flash and jaedong both rose to their dominance." seriously, the term "bonjwa" was crushed by both flash and jaedong, the only difference is, flash had a higher peak, and therefore, jaedong was left in the dark while flash was crowned "god". | ||
|
Mataza
Germany5364 Posts
Look at the length of Saviors dominance. It wasn´t very long, or clear. (Neither was iloveoov) ![[image loading]](http://3.bp.blogspot.com/-tl2E7NIjDME/UBZJ3kRhArI/AAAAAAAAERk/Tft11d56-G8/s640/plot7.png) It seems to me the title of Bonjwa was given to people who discovered a playstyle and dominated with it. That´s why Flash isn´t a Bonjwa, he is the ultimate weapon. | ||
| ||
