Billy Claude

Absolutely agree. One of the first things McDermott did was to plug up all the leaks that were occurring regularly from the front office. Also heavy.com is probably one of the least reliable sources on the internet, which is saying something indeed.

Luckily the coaching staff does not agree with the there's nothing to worry about attitude. That attitude gets you eliminated early on the playoffs. I am almost certain that cleaning up the sloppy turnovers will be a main focus of the coaching staff over the bye week. They have already started addressing the short yardage issues by bringing back the qb sneak. Who knows? They might even be trying to figure out how to cut down on tip passes.

Very short memories on the board. Elam himself said he had an F- game against the Steelers which was only a week and a half ago. Right now Gardner is the DROY favorite at +300 with Devin Lloyd (JAC) and Kayvon Thibodeaux (NYG) at +600 and Tariq Woolen at +800. Elam is tied for 11th best odds at +2500. There are 3 CBs (Gardner, Woolen and Jack Jones (NE)) and one safety with better (Jalen Pitre (TEX)) better odds than Elam.

In Gaelic football, it actually works the other way, 3 points below the crossbar (a goal) and 1 point (a point) above the crossbar. The goal posts are taken to extend upwards forever. They only use cameras to decide on whether a point is scored -- not goals. Actually FGs would be more exciting if the NFL gave more points for lower field goals. That would increase the probability of a FG being blocked and require much more precision and touch from the kicker.

In Gaelic football it is only used to decide if a point is scored, that is if a ball crosses above the crossbar and between the goalposts. It would be similar to deciding whether a field goal is good or not.

Mahomes only started one game in 2017 so the record is 24-13 not 26-14. This is just for the sake of accuracy since the record is still obviously is still very good and does not hurt your argument. Yes, I already noted in my original post that if you take out the 2020 season, the Bills would get much worse. I just thought that it was interesting that most of the KC amazing record came from one season. As I said in my follow up post to your response, I am sympathetic to your argument. The other data you present certainly indicates the Bills are not as good as they should be in these situations. I feel that the Bills bad record in close games is due to a large extend on their inability to convert on short yardage (worse in the NFL). In both the Steelers and Jags game, if they had been able to convert on one of multiple short yardage situations they would probably have won the game. I don't think this is a random thing and was a direct result of trying to go cheap on the Oline.

I thought we were comparing coaches not QBs and Alex Smith was not a bad QB. I am somewhat sympathetic to your argument since I feel the Bills short yardage issues contributed a large part to losing the close games last year.

KC being 26-12 in close games does not seem right so I went back and checked the numbers. Counting any game decided by 8 or less as close, I find that KC is 26-18 in close games since the 2017 season. If you include playoffs they would be 28-21. So KC has been very good in close games but not as good as you claim. Perhaps it was a typo? The regular season corresponds to 59% win rate in close games versus 73% overall. However, a bit closer dive into the data shows that almost all the difference occurred in 2020 when they went an extraordinary 8-1. If you don't include that year they are 18-17 so basically 500 in close games. Of course, if you don't include 2020, the Bills record in close games would also get much worse. I think the conclusion is that teams that are win over a long period of time is because they win a lot of blow outs, not because they win an extraordinary number of close games. Of course, on a single season basis, being lucky may be enough such as with the Bengals last year. For people who are interested, this is the data KC by year, close game record + overall record 2017 3-5 (10-6) 2018 5-4 (12-4) 2019 5-4 (12-4) 2020 8-1 (12-4) 2021 4-3 (11-5) 2022 1-1 (3-1)

If you count a first for Diggs you also need to count a third for Kelvin Benjamin as a day 2 investment.

Thanks for the clarification. Seems much more reasonable.

A 33-23 record when trailing at the end of the 3rd is not likely since I would guess 85-90% of the time, the team winning after 3 quarters wins the game. I like data so I was curious what the actual record is. Here is the Bills' record under McDermott when trailing, tied, and leading after 3 quarters: TOTAL trailing 4 - 25 (0.139) tie 4 - 2 (0.667) leading 44 - 6 (0.880) Here is the breakdown by year: 2017 trailing 0 - 6 tie none leading 9 - 1 (L = CIN) 2018 trailing 1-9 (W = DET) tie 1 -0 ( W = JAC) leading 4-1 (L = NYJ) 2019 trailing 3-4 (W = NYJ, MIA, PIT) tie 1-1 (W = TEN, L=NYJ) leading 6 - 1 (L = NE) 2020 trailing 0-3 tie 1-0 (W = NE) leading 12 - 0 2021 trailing 0 - 3 tie 0 - 1 (L=JAC) leading 11-2 (L = PIT, TEN) 2022 trailing none tie 1 - 0 (W = BAL) leading 2 - 1 (L = MIA)

Definitely a made up stat. The Bills are 52-33, not counting playoffs under McDermott. If they went 33-23 that means 60% of their wins came when trailing at the beginning of the 4th.

They definitely use to be able to do it. I couldn't find the data for 2020 but In 2018 and 2019 Allen was 23 out of 24 QB sneaks which was the best in the NFL https://billswire.usatoday.com/2020/06/08/pro-football-focus-josh-allen-qb-sneak/ This was behind Russell Bodine and Ryan Groy as center in 2018 and Mitch Morse in 2019. One hopes that this season's O-line is better than the 2018 o-line. That season Vlad Ducasse was the Bills' highest rated interior lineman according to PFF. Also the refs/NFL is out to get us.

How is this any different from this board after a lost?

Encouraging losses are only for teams in rebuild mode (i.e., the previous lost against Miami in Allen's first year), but I agree that this lost is not discouraging, as long as any injuries are not serious and they bounce back next week. At least it puts to bed all that 17-0 nonsense.

They seem to be scared of calling QB sneaks since last year's Titans game. Weird, since I think Allen was something like 29 out of 30 before that. I personally was hoping for three QB sneaks in a row, if necessary, at the end of the game.

Looking at the Jaguar game, the Bills had a 1st and goal from the 3 that eventually became a field goal, a third and 3 that ended up as an interception and a third and 2 which led to -1 yard run and fumble by Allen. So not quite 1 yard to go but still pretty short yardage. Against the Steelers, they had a failed 3rd and 1 flea-flicker, a 3rd and 3 incompletion, a 3rd and 3 run for two yards which led to a failed 4th and 1. There were times last year that I felt more confident they would convert on a 3rd and 7 than a 3rd and 2 especially before they started running Allen at the end of the year.

I think they mentioned on the broadcast that the Bills were 32nd in the league last year in 3rd and 4th and one. Plus, their short yardage troubles was one of the main reason they lost all those close games last year. It is definitely something that they should be working on.

I do not wish to get into an argument about this again but it is simply not true. The Verros study for QB injuries from 2016-2020 shows 274 injuries out of 98294 dropbacks (not including spikes) or 0.28% injury rate (these are numbers from the study). For non-kneeldown designed runs, Verros gets 21 injuries out of 3264 plays, which is a 0.64% injury rate, so approximately twice the rate of designed passes. It is not clear where to include scrambles but if you include it in dropbacks, that makes it 309 injuries out of 101,558 plays which would correspond to a 0.304% injury rate, still less than half the rate for designed runs. So based on Verros study, on a per play basis, the probability of injury is twice as high for designed runs as for dropbacks. Obviously, if one gets sacked the injury rate is higher but than compared to a run but that is not the valid comparision. Verros does note that there were about twice as many season ending injuries in dropbacks but on designed runs but that is, again, because there are many more dropbacks and also running QBs tend to be younger and recover faster.

Not sure I agree about the actor part -- the ads I have seen has been pretty cringy -- though it might be the script.

Well I appreciate your contributions, no matter how you do it.

Holy cow. I think this must have been asked before, but do you actually sleep? You do all this detailed analysis (which I greatly appreciate and learn a lot from) and I am assuming that you do the same level of analysis for Arsenal and the Premier League. You also seem to have a pretty high position in the UK government bureaucracy . When do you have time to do all this?

Or more likely, Allen has been so obviously good it's been pretty much impossible to ignore unless you are an obvious troll like Skip Bayless or his apprentice, Nick Wright. (I do realize the original comment was mostly kidding.)

Yes, I did read the study you presented and I am accepting the data they present as true. I simply disagree with some of the conclusions and I have explained why. Yes, a sacks and knockdowns definitely more dangerous than a designed run. However, no offensive coordinator designs a play to get their quarterback sacked or knockdown. To compare the injuries per sack to injuries per designed run is unreasonab;le. If I misunderstood the study please tell me how.

How does the data not back up the claims? You said there were 90 injuries out of 5135 sacks -- since sacks are 6.3% of passing plays that corresponds to about 82,000 passing plays. You give a total of 90 (sacks) + 52 (knockdowns) + 23 (scrambles) = 165 injuries in that time span. So 165 out of the 82,000 passing plays resulted in a injury. This means 2 injuries out of every 1000 passing plays or 0.2%. You quoted injuries at 0.6% per running play so that says, per play, one is three times as likely to be injured on a designed run than a designed pass. It is very straightforward math.