In recent weeks we virtually ceased posting on twitter as well as blogging out betting selections on this site. We assure you, we did not during this period realize that we have a gambling problem and swear off games of chance. Instead, we have been toiling away designing (with the help of some friends) and testing a logistic regression model aimed at improving the accuracy of our NBA forecasts against the spread.
The inputs into our regression model are the same ones we have relied on since the start of the NBA season: two independent routines that each simulate every play of every NBA game several thousand times, and a power rankings factor based on differences in teams play at home versus on the road.
Since going live on February 27, the model has performed admirably, racking up a 60% win rate on a record of 72-47-2.
Of note, the model has been mediocre -- at best -- at identifying home team winners versus the number, but has been magical in highlighting road teams likely to cover. However, we have identified sub-factors that (hopefully) will help us to weed out some of the noise that resides in the home teams selections.
For today, the model highlights the following four plays on AWAY teams:
Portland Trailblazers +2.0 @ Washington Wizards
The Blazers are 7-3 versus the number in their last 10 out, including covers versus several Western Conference contenders (i.e. the Spurs, Clippers, Mavericks and Rockets). We see no reason to expect a letdown tonight versus an erratic Wizards squad that, though they beat the number in each of their last three games, covered the spread facing decidedly lessor competition than they will face tonight.
In labeling Portland two point underdogs, Vegas appears concerned about the Blazers having to play two games in as many nights (the Rose City crew knocked off the Raptors Sunday). However, as the Blazers are 8-5 ATS in games played on zero days rest, this worry appears to us to be overblown. Further, Portland only played 3 games in the week leading up to yesterday's kick off of the current five-game road trip.
Philadelphia 76ers +8.5 @ Boston Celtics
The 76ers have played a bit better of late. In February, Philly managed a winning record of 6-4 ATS and are 4-4 so far this month. Further, the team from The City of Brotherly Love has covered four of their last five out and are 4-0 ATS recently following a loss.
Unfortunately for the Sixers though, the Celts too are playing some of the best ball of their season, covering in their last four contests and in seven of their last 10. However, in head-to-head match-ups between these teams, the road team is 6-2 in the last eight meetings.
We think 8.5 is too much for Boston to lay for a more confident and competitive Sixers unit.
Denver Nuggets +12 @ Memphis Grizzlies
There have been eight occasions this season where Memphis has been cast as double-digit favorites. Of those instances, the Grizz secured the win versus the number once. More broadly, Memphis has covered only once in their last ten games.
Meanwhile, the Nuggets are 7-1 in their last eight games, and have bested the spread recently against the likes of winning teams such as New Orleans, Golden State, Atlanta, San Antonio and Milwaukee.
We see Memphis, the better team, winning outright by a margin of 4-7 points , but do not anticipate a rout, as this is not the Grizzlies style (Memphis' average margin of victory at home is less than 7 points).
Oklahoma City Thunder +6.0 @ Dallas Mavericks
In Kevin Durant's injury-induced absence, Russell Westbrook has authored a wonderful push to propel the Thunder into the playoff picture. The Thunder are 8-4 straight-up (and 7-5 versus the number) since their primary superstar went down in late-February. However, PF, Serge Ibaka has now missed the last two contests, and will miss several more (including tonight's game) as he is reportedly set to undergo a procedure to relieve persistent soreness in his knee. Ibaka has become a more than proficient jumpshooter in recent years, however, his team is likely to miss his defensive prowess tonight as the Thunder face the third best offense in the NBA (Ibaka sports a defensive RPM of 3.56 and an offensive RPM of 0.92).
FiveThirtyEight.com's RPM-based power-ratings make Dallas to be one point better than the Thunder (assuming Ibaka is in the lineup). Based Ibaka's and backup Forward Nick Collison's respective RPM's, average minutes played and the Thunder's and Mavs' average number of possessions per game, subbing Collison in for Ibaka is estimated to result in somewhere around three points of detriment to OKC's spread. Thus, one quantitative assessment of likely personnel shifts calls for Dallas to win by 4 this evening. Compared to the current betting line, we still see value betting Oklahoma City and taking the six points.
Happy betting!!