mramefa

A couple quick corrections not that relevant to my main point: Vegas does ensemble models (not a model) to create ranges of probabilistic predictions, and taking into account how people usually bet they aim for an arbitrage position. They also don't stick to the point spread, and do Bayesian updating based on the bets they receive. Layman's overview of the salient point: the points spread Vegas sets is based on where they think they can make money, given the actions of irrational betters. Just because Vegas has a -7 spread on the Bills doesn't necessarily mean that Vegas thinks the Pats have a >50% chance of winning by at least 7. Rather it means that Vegas thinks it will get action on a -7 point spread where they make a profit no matter what.

Vegas doesn't base the points spread on a single detailed hypothetical scenario...

"Josh Allen has a losing record as a starter." - fredtoast Never let facts get in the way of homerism.

2 RBs is fine for us. If we have to, we can put four receivers out there, and let Allen do the running as our 'check-down'.

1.6M dead cap hit if they cut him after this year. It's a calculated risk for a team with no good TE options last year. With both Sweeney and Knox looking good, and Kroft hurt, it looks like a bad bet. However if neither draft pick worked out and Kroft were looking good, you'd say Beane was a genius. Overall, a smart risk to take, with pretty low downside.

https://www.barstoolsports.com/boston/pmt-9-20-gardner-minshew-warren-sharp-nfl-week-3-preview-mr-portnoy “I was pretty pissed so I started looking around at what I could do and what my options were. The only thing I could do would be to get a medical redshirt,” he said. “So I get an idea. I go home, I grabbed a bottle of Jack Daniels, and I grab a hammer. And I go back in my room, I take a pull of Jack Daniels and put my hand down on the table and boom, boom, boom! One, two, three, and hit the hell out of my hand, dude!”

So despite all the upgrades we are expected to finish 26th, vs 24th last year. Seems reasonable...

https://www.bloomberg.com/news/features/2018-05-03/the-gambler-who-cracked-the-horse-racing-code If you can upload the video to youtube I would appreciate it

When Darnold got drafted, a jets fan I know from NJ texted me with this one

At least one person in Indianapolis is now very happy with Luck's decision to retire.

If Shady had the market value to be worth a trade, then he wouldn't have had to take a pay cut after hitting the open market (let alone a large one like he did). The evidence is there in retrospect that Beane couldn't have done better.

Defense heavy in the draft? Three of the top four picks were on offense...

Who could have every predicted this??? Oh wait...

I think just about anybody could do that

At this point Beane could sell snow to an eskimo, and I wouldn't be surprised.

Corey Bojorquez has better stats, almost across the board: Year Team G Punts Yds Net Yds Lng Avg Net Avg Blk OOB Dn IN 20 TB FC Ret RetY TD 2018 Buffalo Bills 8 45 2,028 1,811 60 45.1 39.4 1 4 8 22 4 9 20 137 0

Found it

I would give up some value for Clowney, even though it would be a one-year deal. I'd make the bet that once he's in Buffalo and gets to see the environment, facilities, and fanbase firsthand, that there's a solid chance he'd want to sign a long-term deal afterwards.

He had back-to-back penalties in the previous game against Carolina as well. They showed a close up of McDermott looking unimpressed. Given last year's record of penalty calls, he's not doing himself any favors with those mistakes.

Playing Barkley gives them a much better chance to evaluate WRs and TEs on the verge of making the roster, though.

Today's game: Nathan Peterman 8/8, 41 yards Kyler Murray 3/8, 12 yards

I plotted preseason vs regular season win percentages for 2018, and it really shows how poor of an indicator the preseason record is. Less than 1% of the variation in regular season win percent can be explained by the teams preseason records.

“Genetics”

I'm skeptical of the idea of doing TE by committee, and adding an overflow of relatively low-profile depth. When low-end draft picks and low-interest FAs succeed somewhere, it's because they're a good fit for the system, and get lots of attention from the coaching staff to keep developing. These players need a lot of work, so I don't even think their long-term potential can be reasonably gauged after a couple months of preseason. Consider the following example. Teams A and B have two TEs already on their roster, and want to add one or two in depth. Team A adds 2 new TEs through the draft and UDFA, and gives them attention and the chance to succeed. Team B adds 10, and lets them battle it out for a spot or two. While team B might end up with a better player, the same player on both teams would be more likely to improve and succeed under team A's approach. I think there has to be an appropriate medium for how much competition is ideal. A way to evaluate this might be to look at teams who added tons of competition at one position in preseason, and then look at some measures of players' NFL success. Have teams that have introduced tons of UDFA competition at a position before, ended up with good quality long-term players?