NBA Line Projections – 22 Feb 21

As always, keep an eye on injuries, back to backs, COVID Issues, and players returning from injury. All stats pulled are TEAM stats from the Last 8 games (L8) and don’t account for individual players. So be aware of new absences and players returning from an absence that could skew the L8 Team stats.

I will regularly post updates throughout the day and annotate with a “*”.

*Models have been terrible ATS the last 2-3 weeks. I have had models go through cold streaks, but the performance recently has subpar for more than just a cold streak. I will be making modifications to the models over the weekend to see if I can make them perform better, but will not be posting the models I am experimenting with. If I can get significant improvement I will move forward with the updated models.

Advertisements

Records

Here is how the models have been performing.

ATS Record

O/U Record

Top Performing Models

Based on Average Win % between ATS and O/U.

  1. Neural Net – 55%
  2. Random Forest – 54%
  3. Ada Boost – 53%
  4. kNN – 53%
  5. Basic Model – 52%
  6. SVM – 48%
  7. Linear Regression – 48%
  8. SGD – 46%
Advertisements

Summary of Projections

ATS Projections

O/U Projections

Advertisements

Player and Team News

Teams on a back to back: Thunder (Cle to Okc).

Bulls/Rockets: Markannen is still out for the Bulls, but he has missed enough games that his absence will be reflected in the stats. Oladipo is doubtful today and appears to be on track to miss his 4th straight game. His absence will be partially reflected in the stats. Cousins is also questionable, if he can’t go his absence will not be reflected in the stats. Wood is still out for the Rockets, but has missed enough time that his absence will be pretty much fully reflected in the stats. The Rockets are 0-7 in the last 7 games without Wood, averaging 104 PPG and allowing 119 PPG. The Bulls are 5-5 straight up, 4-5-1 ATS, and 5-5 to the over in the last 10. The Rockets are 2-8 straight up, 3-7 ATS, and 5-5 to the over in the last 10. Chicago beat the Rockets last time these two met, 125-120.

*Demarcus Cousins is now out for the Rockets.

Grizzlies/Mavs: Dillon Brooks is out for the 3rd straight game for the Grizzlies, his impact will not be reflected in the stats. The Grizzlies have average 103 PPG in the 2 games brooks has missed, compared to the 111 PPG season average. The Grizzlies are 4-6 straight up and ATS in the last 10, and 7-2-1 to the over. The Mavs are 5-5 straight up, 2-8 ATS, and 8-2 to the over in the last 10 games.

Blazers/Suns: No major injuries between either team that isn’t accounted for in the stats, unless a player from either side is a late scratch. These are 2 of the hottest teams in the NBA, both going 8-2 straight up and ATS in the last 10 games. The Blazers are 6-4 to the over, while the Suns are 7-3 to the over in the last 10 games.

Heat/Thunder: Tyler Herro is a game time decision for the Heat, if he can’t go his absence will not be reflected in the stats. Al Horford has not played the second game of a back to back all season. The Heat won the first meeting between these two teams, 118-90. The Heat are 6-4 straight up, 5-5 ATS, and 4-6 to the over in the last 10. The Thunder are 3-7 straight up, 5-4-1 ATS, and 4-6 to the over in the last 10 games.

*Horford is officially out for the Thunder.

Hornets/Jazz: Devonte Graham is out for the Hornets for the 5th of the the last 8 games. His impact will be partially reflected in the stats. The Hornets are 6-4 straight up, 6-3-1 ATS, and 7-3 to the over in the last 10. The Jazz are 9-1 straight up, 8-2 ATS, and 5-4-1 to the over in the last 10. The Jazz won the last matchup between these two teams on 5 Feb, 138-121.

Wizards/Lakers: The Lakers are still without AD and Dennis Schroder. Schroder has missed 2 games and his impact will not be reflected in the stats. AD has missed 5 of the last 8, his impact will be partially reflected in the stats. Without both players on the court the Lakers are 0-2 and the offense has struggled, failing to score 100 points in both contests. The Wizards are 6-4 straight up and ATS, and 3-7 to the over in the last 10. The Lakers are 7-3 straight up, 4-6 ATS, and 5-5 to the over in the last 10 games.

Advertisements

Sharp Report

Early Money

Not a lot of betting data is available this morning.

O/U – Some Sharp money has come in on the Bulls/Rockets Over (226.5).

Mid Day

O/U – Some Sharp money has come in on the Heat/Thunder Under (213.5) & Hornets/Jazz Under (228.5).

Late Afternoon

ATS – Some Sharp action has come in on the Mavs (-4.5).

Advertisements

Model Projections

1. Neural Network (NN)

Record

Record: 51% ATS, 57% O/U.

ATS
O/U

Model Projections

How to read the projections: The model type is at the very top. Matchups are denoted by the alternating white/green pairings, away teams are on top, home teams on bottom. The model analyzes the matchup and projects the home and away teams score (Proj Score). The difference between the home teams score and the away teams score give us what the model says the line should be (Proj Line). The “Line” column is what the Vegas line is at the time I run the model. The “Line Diff” is the difference in the projected line and the Vegas line. A positive line diff means the projected line is in the away teams favor compared to the Vegas line, negative means the projected line is in the home teams favor. The “Cover Prob” uses a normal distribution and the teams variance to project each teams probability to cover the listed Vegas line. Same thing for totals, “Proj Total” is the sum of the projected scores, “Line Total” is the listed Vegas total for the game, the difference between the two, and the probability to go over or stay under denoted by “O” and “U”.

Betting Edge

If you don’t understand what you are looking at, I recommend reading my post about betting tips. The percentages show the betting Edge, which is the cover prob (from above) minus the implied probability (-110 odds implied prob is 52.4%). If a team has a 92.4% chance to win and a 52.4% implied probability (or -110 odds), the Edge is 40%. The Edge alone doesn’t mean you should blindly bet it. To quantify, >35% is great value, 20-35% is really good, 10-20% is decent, <10 is ok value, blanks are negative value. ML is moneyline.

Advertisements

2. Random Forest (R.F.)

Record

Record: 49% ATS, 60% O/U.

ATS
O/U

Model Projections

How to read the projections: The model type is at the very top. Matchups are denoted by the alternating white/green pairings, away teams are on top, home teams on bottom. The model analyzes the matchup and projects the home and away teams score (Proj Score). The difference between the home teams score and the away teams score give us what the model says the line should be (Proj Line). The “Line” column is what the Vegas line is at the time I run the model. The “Line Diff” is the difference in the projected line and the Vegas line. A positive line diff means the projected line is in the away teams favor compared to the Vegas line, negative means the projected line is in the home teams favor. The “Cover Prob” uses a normal distribution and the teams variance to project each teams probability to cover the listed Vegas line. Same thing for totals, “Proj Total” is the sum of the projected scores, “Line Total” is the listed Vegas total for the game, the difference between the two, and the probability to go over or stay under denoted by “O” and “U”.

Betting Edge

If you don’t understand what you are looking at, I recommend reading my post about betting tips. The percentages show the betting Edge, which is the cover prob (from above) minus the implied probability (-110 odds implied prob is 52.4%). If a team has a 92.4% chance to win and a 52.4% implied probability (or -110 odds), the Edge is 40%. The Edge alone doesn’t mean you should blindly bet it. To quantify, >35% is great value, 20-35% is really good, 10-20% is decent, <10 is ok value, blanks are negative value. ML is moneyline.

Advertisements

3. Adaptive Boosting (Ada Boost)

Record

Record: 54% ATS, 51% O/U.

ATS
O/U

Model Projections

How to read the projections: The model type is at the very top. Matchups are denoted by the alternating white/green pairings, away teams are on top, home teams on bottom. The model analyzes the matchup and projects the home and away teams score (Proj Score). The difference between the home teams score and the away teams score give us what the model says the line should be (Proj Line). The “Line” column is what the Vegas line is at the time I run the model. The “Line Diff” is the difference in the projected line and the Vegas line. A positive line diff means the projected line is in the away teams favor compared to the Vegas line, negative means the projected line is in the home teams favor. The “Cover Prob” uses a normal distribution and the teams variance to project each teams probability to cover the listed Vegas line. Same thing for totals, “Proj Total” is the sum of the projected scores, “Line Total” is the listed Vegas total for the game, the difference between the two, and the probability to go over or stay under denoted by “O” and “U”.

Betting Edge

If you don’t understand what you are looking at, I recommend reading my post about betting tips. The percentages show the betting Edge, which is the cover prob (from above) minus the implied probability (-110 odds implied prob is 52.4%). If a team has a 92.4% chance to win and a 52.4% implied probability (or -110 odds), the Edge is 40%. The Edge alone doesn’t mean you should blindly bet it. To quantify, >35% is great value, 20-35% is really good, 10-20% is decent, <10 is ok value, blanks are negative value. ML is moneyline.

Advertisements

6 comments

  1. Is the Last 8 stats something that you weight, or is everything based on the last 8 games only?

    If so, is that the only sport you use that particular criteria with?

    Like

  2. The training data leverages season stats and season data. Todays scores use just last 8 game stats to project the scores. NFL uses a weighted system with the most recent 3 games weighted heavy and season weighted less heavy. Mlb is the same as basketball except it uses last 10 days (typically 7-9 games).

    Like

    1. What about the NHL? Last 8 games as well?

      Also, thanks for the quick reply. I have built my own Pure Points models for all major leagues and always check to see how they match up with your projects. Appreciate all the work you put into your models.

      Like

  3. I did mis-grade the Kings pick, good catch and thank you. The consensus stuff is something i have to do manually and I do make mistakes from time to time. The thunder/cavs game did go over listed 216.5, it was 117-101.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s