SDQL last seasons - General discussion

80% of my super sdql trends with hundreds games behind proven year after year for nice overall 56-60% real profit for five consecutive seasons are broken last two seasons. I quit last season in december with around 50% win rate cause I smell something fishy in almost all of my systems. This garbage was really new to me since I discover these amazing trends. This ridiculous three point shooting with no defense... hundreds of field goal attempts per game and crappy tanking since game 1 ruin the whole perspective in the league. The Warriors are the only team that cover systematically games with some "new" trends aspects like assists, assist to turnover ratio and basket assisted percentage. The other teams are complete garbage that want to copy their game. You need to find a way to broke that style because you just can't copy the best... you will be forever behind them. I wonder what new angles we need to research to find good trends that will work as long as everybody want to be the "new Warriors"

Just see that:

Teams with 35 or more three point attempts last seasons games.

2007 - 16
2008 - 14
2009 - 17
2010 - 14
2011 - 17
2012 - 41
2013 - 54
2014 - 93 (almost double to previous season results)
2015 - 177 (almost double to previous season results)
2016 - 333 (almost double to previous season results)
2017 - 514 (almost double to previous season results)
2018 - ???

2007-2012 were the golden ages for betting on NBA. The game was more predictable. Since then... every season change the game so fast so you just can't adjust and find good angles. Teams in games with 95 or more field goal attempts in 2011 were 71... last season 278

Come on fellas. Let's see if this topic will bring some sdql experts or even beginners to discuss what angles we need to research for this season or we just need to stay away for a while and wait this progression to stop and become something normal so we can find some proven queries which will actually work for the next seasons. The game just makes big steps in new direction and I'm not sure there is hidden profit above the table right now.

Sorry for the English. Last months I didn't practice it at all.

they play "well" vs non 3pt shooting teams. seems like just noise at a 100 game sample. used to know how to find the signifigant sample size needed ages ago :) im sure 100 isnt it for this level of variance though

Damn your second post broke the topic. Maybe some administrator could fix this. My overall point was that last year Rockets did it... this year there will be another 2-3 teams and these numbers will increase for sure... and that just ruin many queries with some criteria that are affected by this insane shooting. Overall the league shoot more, prefer very high tempo with no defense... many players play just for stats and are actually very soft. It's poor product with no true competition. Maybe I will focus on my still working queries and will move on from there. Maybe I will focus more on rest indeed.