The four teams many tabbed before the season to meet in their respective NBA conference finals are set to square off, and while the two in the West got there in predictable fashion, the Cavaliers and Celtics look a lot different than they did when they began the season. Cleveland traded nearly its entire roster at the deadline, and Boston had to overcome an absurd number of injuries, including the loss of stars Kyrie Irving and Gordon Hayward.
Each series was simulated nearly 50 times in "NBA 2K18," with accurate rosters and updated distribution of playing time with anticipated rotations, in order to get the best idea possible of who will reach the NBA Finals.
The "NBA 2K" series has proven adept at predicting outcomes in the past while identifying areas of focus to watch. This year's game already deserves credit for being right on six of the eight first-round series, including correctly predicting the Pelicans over the Trail Blazers and nailing exactly how the Thunder's season would turn out post-Carmelo Anthony trade.
Now, "NBA 2K18" is ready to predict the conference finals...
Western Conference finals
Warriors over Rockets - 61 percent (expected outcome 4-2)
MVP: Stephen Curry - 29 points, six assists per game
Kevin Durant (GS) - 23 pts, 6 reb, 5 ast
Draymond Green (GS) - 12 pts, 8 reb, 10 ast
Klay Thompson (GS) - 19 pts
James Harden (HOU) - 26 pts, 4 ast
Chris Paul (HOU) - 20 pts, 9 ast
Clint Capela (HOU) - 8 pts, 9 reb
Eric Gordon (HOU) - 15 pts
With these two teams, it's not surprising that 3-point shooting often were key to who won the game. The team that made the most 3-pointers at a higher percentage won over 80 percent of the time. The majority of the time that there was a large discrepancy in 3-point shooting, the games ended in double-digit blowouts.
Harden struggled at times shooting from the field. Too often he had lines like 6-of-18 and 5-of-16, and games where he missed all or most of his 3-point attempts. He still put up generally good numbers despite that, but his poor shooting tended to correlate with Rockets losses. Capela, who is one of the players who has risen the most in "NBA 2K18" ratings, was a bit disappointing, averaging below his regular-season numbers in points and rebounds.
As would be expected with the Warriors, every game featured a different hero, much of the time with one or more of their stars taking a back seat in the process. There were games in which Curry exploded and Thompson was virtually invisible, but then Thompson would go off for 40 next time out. Sometimes Durant did it all, and then Green would put up impressive triple-doubles. JaVale McGee also proved productive when he got some extended run.
Home-court advantage looks to be crucial for closing out the series. The Warriors won 86 percent of the closeout Game 6s, while the Rockets won 83 percent of the Game 7s. The first team to steal a game on the road will likely move on to the Finals.
Eastern Conference finals
Cavaliers over Celtics - 95 percent (Expected outcome 4-1)
MVP: LeBron James - 27 pts, eight rebounds, eight assists per game
Al Horford (BOS) - 19 pts, 10 reb, 6 ast
Terry Rozier (BOS) - 19 pts, 6 ast
Jason Tatum (BOS) - 17 pts, 5 reb
Kevin Love (CLE) - 16 pts, 8.5 reb
The "NBA 2K" engine has an understandable bias toward James. A team of James and all the lowest-rated free agents in "NBA 2K18" was found to be better than five actual teams, and the Cavs were given a 100 percent chance to advance out of the first round. Meanwhile, the Celtics have been doubted by everyone, and that's happening again here.
Even in games it lost, Cleveland almost always dominated statistically. The Cavs won crucial statistical categories such as points in the paint, bench points, rebounds, assists, steals and turnovers on a consistent basis.
When the Celtics had success, it was often due to a large discrepancy in free-throw attempts in their favor. That suggests they need to be aggressive, get easy points from the line and get opponents in foul trouble; otherwise, they will struggle to keep pace with LeBron and Co.
The most common result was Cavs 4-1. The series ended in sweeps just 12 percent of the time and went seven games just 15 percent of the time. Maybe this is just another opportunity for the Celtics to prove people (and artificial intelligence) wrong.
The views on this page do not necessarily reflect the views of the NBA or its clubs.