Week 6 Newsletter and Picks
So classic that our models follow up their best ever performance in week 4 with that. There is a whole lot of bad to break down from our week 5 picks so it’s time to make good on our promise to be transparent about our down weeks. The Sunday spread picks went 3-4 meaning we needed the Packers to cover 1.5 points Monday night just to get to .500 on the week. Needing something to go our way just to get to even (and still a -unit week) is never a good sign. Vegas sets these spreads to make them 50-50 picks, or a coin flip. Of course, you won’t get even-money picking a spread which is why you bet $10 to win $9.09 plus your money back. This is where the profit line of 52.4% on spread picks comes from and also why the house always wins unless we have something to say about it. This week, we did not. Of course the Packers went on to lose by 3 and we were left with a 3-5 record on the week. Not good. Over the last four weeks we’ve seen an interesting trend where we have a big +unit week, followed but a smaller -unit week. As long as the +’s continue to be bigger than the -’s we’re ok with that. We hope that this trend continues in week 6 as this would mean we’re in for another good one. Here is the breakdown of the Orb’s models against spread and moneyline picks:
Spreads:
Colts +2.5 ✅
Patriots +1 ❌
Ravens -4.5 ❌
Texans +1.5 ❌
Cardinals +3 ❌
Jets +2.5 ✅
49ers -3.5 ✅
Packers +1.5 (MNF) ❌
All spreads: 3-5, 37.5%, -2.273 units
Moneyline:
Commanders (TNF) ❌
Dolphins ✅
Bills ❌
Colts ✅
Ravens ❌
Texans ❌
Lions ✅
Cardinals ❌
Chiefs ✅
49ers ✅
All moneyline: 5-5, 50%, -2.353 units
Overall an embarrassing -4.626 unit week. Is that bad? It seems bad. That one hurt and took a good chunk out of the season profits we had going into the week. Our models are still +units on the season and are picking 58.6% against the spread. Here is how the season totals were impacted by week 5:
We are burying the game ball that was week 5 and moving on to week 6. Let’s get back on track and back to +unit ways with an awesome Sunday. We are off to a great start as our paid subscribers got a double winner on Thursday Night Football with the Orb taking the Chiefs -10.5 and moneyline. Here is what the Orb predicts is going to happen this week:
Spreads:
Chiefs -10.5 (TNF) ✅
Ravens -4
Commanders +2.5
Dolphins -13.5
Patriots +2.5
Lions -3
Bills -15.5
Moneyline:
Chiefs(TNF) ✅
Ravens
49ers
Commanders
Dolphins
Vikings
Jaguars
Lions
Rams
Eagles
Bills
After favorites have started covering more in recent weeks, our models are backing them to continue to do so in week 6. The models are not afraid of the big spreads on Sunday as Vegas is expecting more blowouts. We clearly do have a bug in our code however, as our models are once again identifying my hometown New England Patriots as a real professional football team. We are far and away the worst team in the league and I don’t know what the numbers see in us this week as we take on the Alibaba version of our franchise in the Las Vegas Raiders.
One of the points of the Orb is to take the human element out of picking games and let the numbers do the talking. Let’s see what they have to say. Comment below your favorite picks this week and let us know if you’re tailing or fading the Orb!
Data Nerd Bonus Content!
For my fellow data nerds out there, we like to be as transparent about our modeling strategy as we are about our results of this project. We are really excited about our multi-model Venn Diagram strategy and the early results we have seen from it. However one concern that I have is with a potential multicollinearity issue. We use advanced team stats as our input variables and I am suspicious that two particular groups of data are correlated with each other and could be hurting our model accuracy. I removed one of the groups and ran a week by week test to see how it would have performed so far this season, deleting each week from the training data before re-training the models. Overall the tests went well and showed that our models would have gone 18-12, 60%, +4.362 units in weeks 2-5. This would be a marginal improvement on how our models have done so far, but in this game every percentage point matters. I am going to keep an eye on this moving forward and if I see enough evidence to do so, I will remove the group of input variables from our models.
- Team Orb Analytics
.
.
.
.
.
.
DISCLAIMER:
The information provided on this website is for informational purposes only. It is not intended to be gambling or financial advice, and should not be relied upon as such. We are not responsible for any actions or decisions taken by readers based on the information provided on this website.
The picks and predictions provided on this website are based on our own research and analysis, and are intended to be used for entertainment and informational purposes only. We do not guarantee the accuracy or completeness of the information provided, and we are not responsible for any losses or damages incurred as a result of using this information for gambling or other purposes.
By accessing and using this website, you acknowledge and agree to the terms of this disclaimer, and you assume all risks and liabilities associated with your use of the information provided on this website.