Saturday Feelings Week 9
Alright, its Saturday again and I am feeling a bit nervous for this week. Coming off a tough week, it’s always nerve wracking coming into the next week as there is potential for people to see two tough weeks in a row and immediately think the analytics are wrong.
I am hoping that doesn’t happen, but at the end of the day, I really don’t have control over that. Hoping for the best!
Week 8 in Review:
Really not many pros to talk about this week. The model went 6-8 again on spread predictions and that is not a good week by my standards. Honestly, it is a bad week. On average you would expect to go 50/50 just guessing, so to go worse than that with analytics behind it is really frustrating. But hey, prediction error is inevitable and the predictions aren’t a recipe to make millions of dollars and I’ve always been transparent about that.
Some highlights this week were definitely the closeness of the games I got correct. The Patriots-Bills game was a good one for example. The model predicted the score to be 21-22 Buffalo, and the final score was 21-24 Buffalo. Additionally, the model was correct in all 5 of the “star rated” games, which actually thinking about it, is pretty cool.
Unfortunately, the biggest loss was the packers losing to the Vikings… this game had me thinking about so many things. I saw a comment from someone on the instagram page that said, “vegas called”. And I had to fall into that rabbit hole… I saw a DraftKings statistic that said 91% of the bets placed last Sunday were for the Packers. 91%!! That is a TON of money. If the Packers would have lost, Vegas would have lost a TON OF MONEY. One could (and should) make the case that Dalvin Cook came out and performed to a really high level and that is why they won. But the money behind these games get you thinking for real.
This week, I will reiterate, I am nervous. So many people at this point look at the predictions and honestly it brings me anxiety to post sometimes. People comment and message really stressful things lol. Like some people will message me saying “I’m putting $1000 on this, you better be right!” And I’m just here like… I have always been transparent and made it clear that these predictions are based on statistics, but they are by no means guaranteed. At the end of the day, statistics is powerful but there is no way to predict with 100% accuracy. If there was, that would mean I could tell the future and if that were the case I wouldn’t need to be doing this. Anyway, I digress…
My nerves are also coupled with the fact that there are improvements that will be added to the model this week and we should see significant improvements over the next couple of days. That being said, the model should still perform like it has in past weeks, but at the end of the day, there are so many choices between which model to use, which data to use, and those choices always bring about different predictions. So at the end of the day, you have to go with the one that performs best in training and hope you picked the right one!
The only real surprise this week in my opinion is the model’s prediction for the Colts to beat the Ravens. This will be a really good game and both teams are really good. As a fan, I don’t see it happening - I think the Ravens are a much better team. But maybe there is something in the data that is tipping off the stats in the Colts direction… can’t be too sure.
Additionally, the model pretty much has the moneyline predictions for every single favorite this week. While statistically this makes sense, I wonder if there are some things that the model is missing. From a stats perspective, the favorite would always be favored to win. This is just a fundamental statistics problem. If there is one team that is statistically better, it would be favored to win and the stats should support that. Obviously that doesn’t always mean that will happen, but from a stats perspective, yes.
Starting this week, I am not going to be watching the games anymore. I get way too stressed watching at this point and need to find ways to stay calm and just read the final scores in as data points instead of watching and getting frustrated at every play. This started as a lot of fun, but it is slowly evolving into a stressful chore in some ways that I need to work some kinks out of.
Everything I do is for free because of my passion for sports and data science. I always post the predictions on instagram the day before gameday, but if you’d like the advanced analytics and predictions for the games days before everyone else, please check out the Patreon page linked here. In the same notion of this being free, every donation is appreciated if you’d like to help keep operations running, click the yellow button at the top of the page for our PayPal.
Check out our YouTube channel here.
And as always, good luck bettors!