Hello all

Every week you'll see a bunch of picks in our table that are predicated on picking based on statistics. The picks are completely arbitrary, so no opinion is ever used, to the point where if there's no statistical separation between the two teams, we'll simply have 'no pick' for that game.

Last week we explained the concept, so this week we don't have to. That's how it works. If you do want to read a bit more on the stats themselves, I do highly recommend last week's riveting read.

Straight up

The first notable point is clearly that as with week one, an 11-5 record for 'Best Defense (PPG)' has again been good enough to stand at the top of our mini-standings as the best statistic for picking games. Best D was joined at the top by 'best run game', based on offensive rushing yards per game.

Looking at the games where 'best defense' correctly contradicted the expert consensus, and there is one glaring consistency - all of the 5 games involved picking the underdog, and 4/5 were road underdogs (Minnesota the exception). This is interesting and something we'll keep an eye on as we revisit these stats each week to see if there is merit to picking the team that gives up the least points when they're on the road in adverse conditions. I think it's worth being careful not to jump too quickly to conclusions, however, particularly about favorites/underdogs early in the season. The public (who dictate the spread) are still finding their feet with these teams, so they're still going to under-value teams like the Eagles, Vikings and Titans until they have won over a prolonged period. It's too early to say, but I'd suggest that there's a probability that picking based on certain key stats at this point is actually far better than trusting opinions, as a lot of the latter will inevitably be based on preconceived ideas that have developed through the offseason, ie: 'this team will rebound in 2016' or 'this team lost it's QB ergo is terrible'.

In the clutch standings, 'Rushing Yards Allowed' nailed a perfect 5/5, showing the importance of a strong run defense as a complimentary component to not giving up points. So far all of the 3 defensive-orientated stats have fared well, in direct contrast to the mediocrity of offensive stats, none of which have broken 60% on the year.

As for best kicker? Well... it's not been good. Of particular note is how close many of these games have been so far in 2016, and yet having the better kicker has not been a reliable differential between evenly matched teams.

Against the Spread

We have one ATS-specific stat, 'Average points difference', and boy has it started with a bang. Two winning weeks and an overall 59% rating makes it a very interesting ATS stat to follow. I'd expect that stat to improve over the next few months, because basing it on one week is hardly a great sample size, and even two weeks is a stretch. Over time we'll see the average points for each team 'normalize' to the level they're playing at in 2016 and outliers where games put up way more or less than 'normal' will become less decisive in these stats.

The 'best run game' stat would have been a great one to follow with an 11/5 ranking, but worth noting the inconsistency vs week 1 (5-11). On a consistency note, our average points differential is the only stat to have two 'winning' weeks vs the 52.4% base line for beating Vegas over a whole season.

No clutch stats for ATS yet (although that's something to add to the list) and frankly, all games vs the spread are by their nature going to fall into our existing 50-74% definition of a close game. As for the bad performers, well, again don't trust your kicker, and so far this season, don't trust ESPN's QBR stat either, which has delivered the best option in just a third of ATS games. Bad form. Literally.

Next week we'll be back with another breakdown and we should begin to see the development of some of these stats over the first quarter of the season take shape.