Best Squads in TWLB History

[ogron]

So to make any realistic comparison of the Best Squads in TWLB History it would HAVE to be relative.

I've added the vs. Era column to answer your question (within the limits of the methodology.) It's simply the overall team's Kills Per 30 Minutes, minus the season average KP30M. I excluded 6 teams (in red) from the seasonal averages because they dragged it down way too far (For instance, Season 18's actual KP30M average is 83 if I include Warhable, which would give too much of a boost to Pandora/Dice relative to Penetrate/Thunder Season 12 who played in a more competitive season, in case you wanted to compare the most dominant teams statistically across seasons.

[ogron]

Most Dominant Squads within their own Era




I'm not a fan of this particular way to rank squads -- as you can see Penetrate S12, which had ranked 3rd overall in KP30M on the original list, falls to 14th, since they played in the strongest season in the last 7 years.

Using KP30M alone, IMO, is more predictive, because while earlier teams like Pene S12 played tougher regular season schedules, they also played less games in the playoffs. The stratification of the league (read: stacking) and quality of play improving is offset by more playoff rounds vs. more stacked opponents, and those games last longer.

Pandora S18's KP30M doesn't benefit much from destroying Warhable 15-1, even if they get 130 KP30M vs them, as it only accounts for 80 (16 minutes x 5 spiders) of their 1400 minutes played. By comparison, they played the 2 of the most stacked squads ever (Dice S18 and Enemy S18) a whopping 7 games for over 800 minutes across both the regular season and the playoffs, not to mention the 4th, 5th, and 6th best squads in the League were quite good (Value took them to 15-14, for example.)

The predictive value of KP30M is very strong, as you can see the majority of the top teams on the list (in the original post) are either champions or runner ups. I'm too dumb at math to figure out a coefficient, but I'm sure if someone ran it on KP30M relative to winning a title or reaching the finals, it would be quite strong.

[kyler]

I've added the vs. Era column to answer your question (within the limits of the methodology.) It's simply the overall team's Kills Per 30 Minutes, minus the season average KP30M. I excluded 6 teams (in red) from the seasonal averages because they dragged it down way too far (For instance, Season 18's actual KP30M average is 83 if I include Warhable, which would give too much of a boost to Pandora/Dice relative to Penetrate/Thunder Season 12 who played in a more competitive season, in case you wanted to compare the most dominant teams statistically across seasons.

did..did we not make it to the top squads basing squads of all time..?

[ogron]

Stats never tell the whole story. The eye test is needed. Too bad Youtube wasn't around for most of this game's lifespan.

I 100% agree stats never tell the whole story. This has a been a preamble for most of the posts I make, I didn't feel like I had to write the obvious for the umpteenth time.

That said, I ran a Pearson product-moment correlation coefficient, and the relationship between a team's KP30M and their end of season rank (1st, 2nd, 3rd, 4th, etc.) is -0.803. +1 or -1 is almost causal, while 0 means no relationship, so -0.8 is extremely strong. In terms of predictive value it's a very good tool for estimating the strength of a team.

The eye test is generally the best assessment if the eye is knowledgeable on the subject and the viewing is fresh. Given that we have few youtube videos of gameplay and memory is extremely faulty, one major benefit is that statistics don't decay. The downside is that LEGO S16 is ranked a smidgen above Thunder S16, despite the latter being better. Thunder smashed LEGO in the QF, and the only reason their KP30M was lower was because Mattey didn't play all the regular season games, Shayde disappeared mid-season, and Ease missed the SF for a wedding and they lost 20-18. That squad should have had a 90+ KP30M and been in the Championship game, but the statistics give no indication of that.

That is the exception rather than the rule, however, and overall the stats do a good job of indicating the strength of the teams. Again, it's best to separate them into Tiers than consider Team A being better than Team B to account for missing players in certain games, luck, etc. It provides more meaningful data on the whole.

The flipside of the stats incorrectly listing LEGO above Thunder is that it can also demystify some teams/seasons by providing a more balanced and objective assessment. For instance (I have no idea if this is true or not, I never saw Season 12) it's possible people overvalue Pirates since they made the finals and their name is etched in TWL History and that Thunder was a better team and luck/other unfavorable values contributed to their SF victory, and that the stats are looking at the entire season on the whole, not just the result of one game, and are therefore correctly listing S12 Thunder over S12 Pirates, despite the latter finishing higher.

did..did we not make it to the top squads basing squads of all time..?

LOL

[Mighty]

Took the words right out of my mouth but relax, some people rather draw a basketball and subspace images with arrows than actually go outside & search for a job.

don't forget to include the multiple essays a week lmao

[Jones]

stats has always been enjoyable to read. Alot more in the past than now though, happens when you've been here for a while.

A good contribution claus, much appreciated. a little bit overwhelming at times though

[grand]

gay thread, gay post, gay everything

issues with all of this:

1. dicE best basing dynasty of all time, pls recognize this
2. analysis doesn't even separate or analyze TWLB pre-removing the two blocks and post-removing the two blocks. bet most of u noobs dont even kno what im talking about
3. missing a shout out to my #1 homies juice & apt
4. reading this thread reminds me how sad TWL has been the last like 7 years. i miss the golden age

[RoDNeY]

Rofl @ the people trolling claus, guy can do w.e the fuk he wants with his time, if he wants to put it into a game that he loves that we've all played for years then fkin let him
Jeeezuz fuk this forum is full of washed up fuk bois

[ogron]

Appreciate the kind words from Vati/Rainbow/Hot Rod : )

In truth, it doesn't take much time to do -- you can copy/paste the stats from TWL into excel then run a formula, so it's pretty quick.