Mig_22

I said at the beginning that the Sabres and Jack would end up with the same amount of points. I had it at 93. That's close. I'll vote yes.

Mrazek played last night . I assume we'll get McElhinney?

Maybe there is a small debate left. ? The stats are for the 10 games of the streak, not the season. During that time, the Expected Goals For was pretty good (don't have a number handy, sorry). Also, when you look over the season, their xG % (For vs Against) is mid-pack in the league 5v5 and Hutton's overall numbers are about 12th-15th in the league. I'd say we're winning games for a whole lot of reasons including both timely goaltending and a bit of puck luck.

Great point. Most of the uses (and misuses) of the concept tend to be in the negative.

Exactly. I hate to use Toronto as an example but in their dismantling of San Jose last night, this is Micah McCurdy's HockeyViz.com chart of the game. San Jose had the better of the play for chunks of the game even while getting waxed. I think it's been a strength of the Sabres that their bad stretches have been shorter than their good stretches.

Most preseason predictions put them at 82-87 points. "regresssion to the mean" is often misunderstood to mean that a result will get back to an original prediction. More accurately, it is that results will tend to the same average as expected going forward and thus pull the mean lower. So GASabres point is quite right. We can, without taking our current form into account, reasonably expect an 85 point pace for the rest of the year (~59 points in 57 games) which is 95 total. Sure we could miss the post-season. Edit: I don't think that will happen because I didn't buy into the preseason numbers and had their total at 92-93.

But what is missed is Sam contributing his talents to another line. He really does drive play well.

The low-medium-high danger scale used by Perry at Corsica is based on Expected Goals for an unblocked shot (Fenwick). That metric takes into account location, rush/no-rush, rebound attempt or not, and one or two other things. So in 2016 when he created the category, the percentages were as follows: Shooting percentages are up league wide since then so the actual percentages probably are different now but Hutton's somewhat poor save percentage on medium danger shots is important since that will be about 40% of the shots faced. I still love the guy but he is playing solid-to-well, not spectacular. (last 7 starts he has been gaudy good). It's certainly been enough for the Sabres.

I'm happy enough with Hutton and your stat about high danger save % is important. One thing is that his medium danger save % is really sub-par at below 90%. This doesn't invalidate your point at all because saving the tough ones has more benefit sometimes than just a save but the overall body of work by a goalie is important and Hutton is pretty much mid-pack. Now that is a big improvement over 2017-8 Lehner but he is not carrying the team. Damn, goalies are hard to quantify.

The goaltending is better than last year, although about the same as the year before, but still the metrics (save %, goals saved over average) have Hutton right near the middle of the pack. It's not like he's putting up a wall back there. I have a theory that I need to research and would welcome thoughts on. The team still makes the mistakes in coverage and turnovers leading to the high number of slot opportunities that Flagg mentioned. Also, there are periods (sometimes whole periods) when the team runs around its own end and looks terrible. The difference in the defense this year is that those times, while still really bad, do not make up a large portion of games. On corsica hockey, you can see the expected goals in a lot of the Sabres games this year be very tight with the opposition. So the idea is, that they are playing just as poorly as ever sometimes, but for shorter periods of time. Testing that with available data is probably a real chore.

Well, the three lynch-pins to the team's season-long improvement are Dahlin, Middelstadt, and Nylander. The team is hovering at the playoff line with none of the three being exceptional. Assuming they stay healthy and improve, I think it's not premature to say that the team is leaning towards good. McCurdy's projection puts the Sabres at 50% to make the playoffs, Dom at The Athletic has a 30% chance. Both of those models are admittedly speculative about young players/prospects because of lack of data. So given those numbers and the Sabres record right now, would you make an even money (50% chance) bet to make the playoffs? Or would you need 2--1 (33% chance)?

Interesting to see and fun to watch. But there are costs of course. On Domi's second, McCabe was high at the blue line and Rodriguez was left to defend Domi...which he didn't do at all. But still, it is an exciting thing.

at ~20% time tied, that looks like it is the lowest in the league. That's just eyeballing it but no other team's trailing and leading time adds to 80%

That's 8 out of the last 9 periods that the Sabres were as good or clearly better than the opposition. (3rd against the Sharks was a mess). What Buffalo fan doesn't take that? Good times.

He stopped 32 shots from the slot last night. I can't wrap my head around that. But we should get Miller tonight?

Anaheim has a good record but Gibson is carrying them. He has stopped 2 goals more than expected (based on shot location). Even if the stats aren't perfect, he has been absurdly good. And we probably face Miller instead since they played late yesterday. Opportunities.

I agree with a lot of this post. How Housley deploys his centers is a very astute observation and Jack's positioning is most noteworthy. I think I agree that he is going to try Sobotka in the defensive side of the ROR role and I'm okay with it because you're not wasting a good offensive talent when doing that with Sobotka. I don't mind Berglund doing that work either but like you, I haven't entirely formed my opinion on whether the overall structure hurts the offensive production too much. With Jack, I think it doesn't because much of his 5v5 production should come on the rush (in theory). More thinking and watching on this is in order. Cheers,

I read somewhere that they had 3 analysts on the payroll. I don't know if it's a good use of time for them to actually do the grunt work of collecting the standard stats by watching the game instead of buying it from stat collectors and augmenting it with public data and their own targeted work. That's how I would organize it as a database builder. But you might be right and they do all of their data collection themselves. One note though is that that public data now has a LOT more attention on it from writers and the league and that crowd-sourcing will greatly enhance the data quality. And you are very correct that the teams keep all their own juicy stuff and don't release it. Why would they, right?

Yes, the Sabres are bad at shooting. And strangely, I think Jack's one glaring weakness is his lack of shooting accuracy but that is just my eye-test so it may be all wrong. Regarding the accuracy of the stats collected, I am certain that it has improved. Teams now analytics staff, not just one 'nerd' that knows Excel, and those staffs need accurate data to supply good scouting to the coach and GM. I imagine that they can't collect all their own data so they use the league data and some other specific sites and it is in everybody's interest that the data be pretty solid. I'm sure it's not perfect but definitely improving.

Vogl's followup tweet was that Housley said Berglund had some bumps and bruises.