Week 15 Newsletter and Picks
Another week of football and picks to dive into as the regular season is starting to wind to a finish. Week 15 is why the house always wins. I couldn’t find an exact amount but I have to imagine that Vegas the bookmakers are eating good right now. Underdogs absolutely dominated as just 4/16 favorites covered the spread and only 6/15 won outright. That’s right, the majority of games in week 15 were won straight up by the teams getting points. I can’t imagine anyone making it through Sunday with a winning moneyline-parlay but if you did my hat is off to you. Thankfully our models loved the underdogs covering as we had a great bounce-back week against the spread. It wasn’t one for the record books, but the Orb hit 61.1% of its spread picks which is encouraging to see after the slide in recent weeks. From a data perspective, it is also cool to see a week perform right in line with our season record, rather than greatly above or below it. On the other side, you can probably guess by now that our moneyline picks were ugly. If you are new to the Orb, we have a heavy internal emphasis on the performance of our models against the spread, but we do run them for moneyline and need to be just as accountable for these misses.
In contrast to week 12, this week’s performance will go down as an overall moderately heavy loss to our season unit total record but felt like a win with how the Orb fared on spread picks. We can’t go through all that went right and wrong for the Orb without first addressing the elephant in the room: The Carolina Panthers. I cannot figure out why, but for some reason, our models keep identifying them as a real-life NFL football team when they clearly aren’t. I have my own personal thoughts on Bryce Young and the roster as a whole this year but I don’t want to pile on to any Panthers fans reading this. But for purposes of the Orb, I can tell you just how wrong our models have been on them. Our models have picked them to cover 6 times this season, all as underdogs, and have gone 1-4-1 on those predictions. This translates to them costing our models 2.5% percentage points and 3.091 units on the season. All this is to say that I am stepping in and making an executive decision to hard-code out any Panthers picks against the spread this season going forward. Here is how each pick went in week 15:
Spreads:
Patriots +6 (TNF) ✅
Panthers +5.5 ❌
Bears +3 ✅
Raiders +3 🟨
Chargers - 2.5 ❌
Chiefs -1.5 ❌
Cowboys -3.5 ✅
Titans +13.5 ✅
Giants +6 (MNF) ✅
All spreads: 5-3-1, 61.1%, +1.545 units
Moneyline:
Steelers (TNF) ❌
Ravens ✅
Saints ✅
Falcons ❌
Lions ❌
49ers ✅
Chargers ❌
Chiefs ❌
Cowboys ✅
Dolphins ❌
Packers (MNF) ❌
All moneyline: 4-7, 36.4%, - 5.715 units
Overall a -4.17 unit performance from our models. After the Panthers game, our next miss came with Justin Herbert’s season-ending injury. Herbert is now added to a long list of starting quarterbacks lost this year to injury which hurts the product on the field and muddies the team’s data. All credit to Denver though, I think there's a good chance they would have spoiled that pick anyway with how well they are rounding out the end of the season on this playoff push. Finally, we found ourselves on the wrong side of the most talked-about play of the day:
https://larrybrownsports.com/football/kadarius-toney-chiefs-offsides-touchdown/625409
Kadarius Toney lining up offsides brought back what would have been an all-time fun touchdown as the Chiefs thought they had sealed this one. We will never know if Buffalo would have been able to create a last-minute drive to take back to lead, but assuming there wasn’t enough time, Toney’s foot theoretically created a 3.729 unit swing against our record this week. It’s a game of inches and we unfortunately just found ourselves on the wrong end of a close one here which happens. Again, this was a good and important result overall for our models against the spread but I always think it’s more interesting to see what the models got wrong.
In the data that we train our models on, the favorite won the game outright 67% of the time. We only give out picks when all our models agree, which is why the Orb predicts favorites to win way more often than not. This tries to accomplish the goal of giving out winners as well as not giving out losers. This means any pick that one model takes the underdog to win outright won’t be given out if the others don’t as well. I don’t have the exact number but this has filtered out giving some heavy favorites to win that ended up losing the game because one model did not trust their data. But what it also means is that our moneyline picks are prone to weeks like this. When underdogs win outright at a high rate like they did all last week, it will hurt our moneyline totals. Here is how week 15’s results impact our season totals:
Season-to-date results:
Season spread picks: 51-33-5, 60.1%, +13.61 units
Season moneyline picks: 95-41, 69.8%, +5.81 units
An interesting, kind of ugly impact on our season-to-date unit graph but as long as that orange line goes up, I am happy and we are one week closer to accomplishing our season goal for this project. Now let’s keep the spread momentum going in week 15. After a week of underdogs, our models are leaning slightly more on favorites bouncing back on Sunday:
Spreads:
Browns -3
Patriots +8.5
Cardinals +12.5
Bills -1.5
Ravens -3
Moneyline:
Packers
Falcons
Browns
Chiefs
Bills
Ravens
We have picks for Saturday's games and potentially Monday as well for our premium subscribers. If you are interested in who our models like Saturday, or just want to support the two men behind the Orb feel free to sign up for access to our non-Sunday picks for the rest of the season.
One last bit of housekeeping to close out the Newsletter and just something I want to get ahead of. Our goal here from a modeling perspective is to give out picks through data that is representative of the teams that we are going to see on the field. In the same vein as why we didn’t do official picks for week one, we also are not planning on doing official picks for week 18. Most teams either have their playoff positions locked up or are trying to get a better draft position by losing. This results in many teams trotting out backups and resting their key players which means that the data they created all season is not actually what is on the field. Last season for week 18 (and this year for week 1) we still ran the models and posted what they predicted, but with heavy disclaimers about why they may not be accurate or official picks. As of right now I am planning on doing the same thing in week 18 but will keep you posted if we have games that matter enough for both teams to field their full rosters.
Have a great week 15 everyone!
- 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.