re: Week 8 SP+ Picks

quote:

---

That's not entirely true. Saying it's math makes it seem like this concrete thing that yields only correct results. It's data to which he applies his own weights.

I don't put any stock in these, because I don't really believe in relying on predictive models to tell what's going to happen in a game where a star player could have a bad game because he woke up in the wrong mood.

For example, the week 7 pick for UGA/SC was 36-17 in favor of the dawgs. While I understand that was an outlier, shite like that happens all the time.

They are fun numbers to look at, but I put absolutely no stock into them when I'm deciding my picks.

---

It's a probabilistic projection. It's always going to give you what it sees as the single most likely outcome out of a vast number of possible outcomes. That doesn't mean that the particular outcome it forecasts is likely to happen; in fact, they'll barely ever nail the score exactly. It's a median result; if the math is right each team will outperform the projection roughly half the time.