2021-2022 Bruins Season Thread

[Smiling Joe Hesketh]

Swayman is really struggling lately. Last 7 games: 3-4, 3.50 GAA, .880%.

I don't know if it's SSS, just a slump, or hitting the rookie wall. But he hasn't been good. I wonder who starts for the playoffs.

 

[lexrageorge]

Swayman is really struggling lately. Last 7 games: 3-4, 3.50 GAA, .880%.

I don't know if it's SSS, just a slump, or hitting the rookie wall. But he hasn't been good. I wonder who starts for the playoffs.

I think the door is open for Ullmark at this point. There are 9 games left, with 2 back-to-backs, so each goalie probably gets 4 games to audition.

 

[cshea]

To be honest, I think they're going to rotate.

 

[ColdSoxPack]

Nothing in the Globe about tonight's ceremony but this is what NHL.com has. Seems not star-studded:

Members of the 1972 team in attendance will include Don Awrey, John Bucyk, Ken Hodge, Don Marcotte, Garry Peters, Dallas Smith, Rick Smith, and family members of Ace Bailey and John McKenzie. Members of the team, as well as Bruins coaches and staff, will wear limited-edition pins honoring the entire 1972 team.

 

[TheRealness]

To be honest, I think they're going to rotate.

I would guess it’s hot handish, but I feel like they are so equal at this point that you could see them rotate just to give the team a different look. I’ve been very impressed with Ullmark lately. It seems he’s settled in a lot.

Swayman seems tired to me, and I would bet it’s more mental fatigue than physical.

 

[Smiling Joe Hesketh]

Right on cue, Ullmark gets hurt and Swayman goes in and can’t stop a beachball.

 

[Dummy Hoy]

Right on cue, Ullmark gets hurt and Swayman goes in and can’t stop a beachball.

Swayman was thrown into the fire and had to deal with a parade of power plays...he settled in and played well in the third.

 

[Smiling Joe Hesketh]

Swayman was thrown into the fire and had to deal with a parade of power plays...he settled in and played well in the third.

Still gave up 3 goals in one period to lose to one of the worst teams in the league.

If Ullmark is healthy, he's my starter for the playoffs. I don't think Swayman is anywhere near ready. His recent form even before last night was very poor.

 

[cshea]

Ullmark is probably concussed. He is not practicing today, Grosenick is up from Providence. Ullmark didn't get pulled by the spotter but told the staff he wasn't feeling well after the first period.

We'll see. 8 games to figure out the game 1 starter. Regardless of where that lands, I would be sharing the net in the playoffs unless one of them gets white hot.

 

[cshea]

I would say that on the whole, this is what average goaltending looks like. There's ebbs and flows. Swayman was hot after the Rask fiasco, now he's cold. Ullmark started slow, now he's hot but injured.

The bugaboo is high danger save percentage. Ullmark is .814 (37th) and Swayman (39th) among 59 goalies who have played 1000 minutes. Both guys are basically making the saves they are supposed to make but not getting the high danger, highlight reel saves at a rate we're accustomed to (Rask was top 5 in total last 2 year),

They do face very little dangerous chances against. Swayman faces the fewest per 60 in the league, Ullmark is 7th. The Bruins keep everything to the outside. In addition to not allowing many high danger chances, Swayman faces the furthest average shot distance against in the league, Ullmark is 8th. The problem is when the defense does break down, which happens no matter how good your defensive structure is, Ullmark and Swayman aren't bailing them out as much as we'd like. This is essentially what happened last night. At even strength with Swayman in, they gave up an xGA of 0.99. He allowed...1 goal. He didn't stop either of the lasers on the PP which aren't really his fault but that was the difference.

Neither of these guys have NHL playoff experience. They've also never been full workhorses. Ullmark made his 38th appearance last night which is a career high. I don't think either is ready or capable of the every-other-night, possible multiple OT's, grind of a playoff run. Win or lose game 1 or 2, I think the sensible play is to rotate them and keep both fresh. Can always re-evaluate as a series moves on.

 

[j44thor]

Just curious does the xGA account for saves while the team is on the PP? B's have been giving up an inordinate amount of odd-man rushes lately on the PP that xGA seems rather low considering Sway did stop multiple breakaways in the 3rd and a couple odd-man rushes, at least a few of which came while B's were on the PP I believe.

 

[PedroSpecialK]

That's also one of the issues with evaluating performances like last night's in hindsight - B's were clearly the better team, but the game turned on the whims of a washout of a player who is so bad as a ref he kicked Reilly a puck in the course of play last night.

Different discussion for a different thread, but enough is enough. The two-ref, two-linesman system is abominable and is getting worse as the game speeds up. Remove a ref from the ice surface, put him off-ice with authority to stop play on missed calls, and see how it goes.

 

[Ferm Sheller]

but the game turned on the whims of a washout of a player who is so bad as a ref he kicked Reilly a puck in the course of play last night.

I forgot all about this. This is so wrong I don't even know where to start. I think he must know that he was in the wrong, and hopefully he won't ever do it again, but jeesh, can you imagine a ref doing that in OT of Game 7 of the SCF?

 

[PedroSpecialK]

Didn't get far from it in Game 5 v. St. Louis

aaand I'm chewing on rawhide again

 

[cshea]

Just curious does the xGA account for saves while the team is on the PP? B's have been giving up an inordinate amount of odd-man rushes lately on the PP that xGA seems rather low considering Sway did stop multiple breakaways in the 3rd and a couple odd-man rushes, at least a few of which came while B's were on the PP I believe.

No, that's even strength.

In all situations, it was 2.45 xGA Swayman and he allowed 3. On the PP, they allowed 0.31 xGA. The difference was the PP. He allowed 2 goals against vs. 1.15 xGA on the PP and that's the difference. I don't think you can really fault him for either of them, particularly the 5x3 goal but a save on either of those and maybe they win. The first goal against was bad positioning and you'd like to have that one back but it is what it is weird ones happen, he matched the xG. As long as they don't give up more actual every nighthan what's expected, and they don't, they should be OK.

 

[Myt1]

Still gave up 3 goals in one period to lose to one of the worst teams in the league.

If Ullmark is healthy, he's my starter for the playoffs. I don't think Swayman is anywhere near ready. His recent form even before last night was very poor.

I think Swayman’s upside is higher and Ullmark’s positioning is really hit or miss. He’s been hot, but I think he’s generally a worse goalie, and this seems clear by their season long and career stats.

 

[Myt1]

I would say that on the whole, this is what average goaltending looks like. There's ebbs and flows. Swayman was hot after the Rask fiasco, now he's cold. Ullmark started slow, now he's hot but injured.

The bugaboo is high danger save percentage. Ullmark is .814 (37th) and Swayman (39th) among 59 goalies who have played 1000 minutes. Both guys are basically making the saves they are supposed to make but not getting the high danger, highlight reel saves at a rate we're accustomed to (Rask was top 5 in total last 2 year),

Am I reading something wrong? I see Ullmark’s HD SC Sv% as .781 (https://www.hockey-reference.com/players/u/ullmali01.html) and Swayman’s as .808 (https://www.hockey-reference.com/players/s/swaymje01.html)

By comparison, Sheterskin is like .857.

 

[cshea]

Am I reading something wrong? I see Ullmark’s HD SC Sv% as .781 (https://www.hockey-reference.com/players/u/ullmali01.html) and Swayman’s as .808 (https://www.hockey-reference.com/players/s/swaymje01.html)

By comparison, Sheterskin is like .857.

I was looking at Natural Stat Trick. Maybe they define high danger differently.

 

[Myt1]

I was looking at Natural Stat Trick. Maybe they define high danger differently.

Yeah, I figured maybe there was something like that.

Knowing absofuckinglutely nothing about the relative rigor of the two sources, it really would be illustrative of something possibly useless about the stat if two sources were calculating the same thing so differently. Maybe there’s a wrinkle, but I’m chuckling into my drink at the absurdity of us having looked specifically for the same thing and coming to wildly different conclusions based on our sources.

 

[Frisbetarian]

Yeah, I figured maybe there was something like that.

Knowing absofuckinglutely nothing about the relative rigor of the two sources, it really would be illustrative of something possibly useless about the stat if two sources were calculating the same thing so differently. Maybe there’s a wrinkle, but I’m chuckling into my drink at the absurdity of us having looked specifically for the same thing and coming to wildly different conclusions based on our sources.

A lot of the publicly available stats can be quite misleading. The first issue is the data source, which (I believe) is from the NHL and not all that accurate. I mean, it's getting better (when I first started in hockey I had to create a rink effect because some teams shot locations were off so much and so consistently - the Rangers were the worst), but it's still not nearly as accurate as what the more analytically inclined teams use, or what you can get from places like Sportlogiq and Stathletes. There's also a definition issue, in that it's possible NST had a different location for high danger chances than HR (inner slot vs slot maybe?). Then you should consider context, as not all high danger chances are the same. On some the player is alone (breakaway), other times the shot is screened, etc., etc. Sample size of course needs to be considered with a stat like HDC, as the SV% can change quickly in a smaller sample. Therefore, you really should have an idea when data points stabilize and can be used with some confidence (believe it or not, to his credit Cam always asked this question on early XPG data).

Expected goals on ice and expected individual goals are both influenced fairly heavily by a player's OZ starts, which need to be considered when assessing a player. An aside, but it would be kind of cool to test the different XPG models that are publicly available to see which most accurately predicts XPG/XPGA/XPG% vs actual goals on ice. I'm not up to date on public hockey data; I know Natural Stat Trick, Hockey Reference, and NHL.com all have XPG data. Is there anyone else? Am I the only geek who would be interested in seeing how the different public XPG data compares?

 

[TSC]

Not great!

View: https://twitter.com/flutoshinzawa/status/1516085652290359299?s=21&t=oE_xb3naz_KNpM3AzfLCpw

 

[Ferm Sheller]

Not great!

View: https://twitter.com/flutoshinzawa/status/1516085652290359299?s=21&t=oE_xb3naz_KNpM3AzfLCpw

No need to rush them.

 

[burstnbloom]

A lot of the publicly available stats can be quite misleading. The first issue is the data source, which (I believe) is from the NHL and not all that accurate. I mean, it's getting better (when I first started in hockey I had to create a rink effect because some teams shot locations were off so much and so consistently - the Rangers were the worst), but it's still not nearly as accurate as what the more analytically inclined teams use, or what you can get from places like Sportlogiq and Stathletes. There's also a definition issue, in that it's possible NST had a different location for high danger chances than HR (inner slot vs slot maybe?). Then you should consider context, as not all high danger chances are the same. On some the player is alone (breakaway), other times the shot is screened, etc., etc. Sample size of course needs to be considered with a stat like HDC, as the SV% can change quickly in a smaller sample. Therefore, you really should have an idea when data points stabilize and can be used with some confidence (believe it or not, to his credit Cam always asked this question on early XPG data).

Expected goals on ice and expected individual goals are both influenced fairly heavily by a player's OZ starts, which need to be considered when assessing a player. An aside, but it would be kind of cool to test the different XPG models that are publicly available to see which most accurately predicts XPG/XPGA/XPG% vs actual goals on ice. I'm not up to date on public hockey data; I know Natural Stat Trick, Hockey Reference, and NHL.com all have XPG data. Is there anyone else? Am I the only geek who would be interested in seeing how the different public XPG data compares?

i would be interested. The public models are also built somewhat differently, which is why EH and dom can have variance in their evaluation based on the same raw data. I believe EH doesn’t use chance data for the reasons you stated with the public data but instead regresses raw shot data to account for game situation. Dom has said that he uses chance data. The nuance is why when I’m using those models evaluations, I look at all of them. It’s rare that there are huge variations between them so if all of them say player x is performing at y level, it feels reasonable to accept that as indicative of the players performance.

I think regular hockey fans are probably going to live in the gray until player tracking is easily available, the samples that creates are large enough to trust, and the models have been tweaked to push out accurate evaluation. What we have today is worlds better than we had ten years ago but there’s still a ways to go.

 

[Myt1]

A lot of the publicly available stats can be quite misleading. The first issue is the data source, which (I believe) is from the NHL and not all that accurate. I mean, it's getting better (when I first started in hockey I had to create a rink effect because some teams shot locations were off so much and so consistently - the Rangers were the worst), but it's still not nearly as accurate as what the more analytically inclined teams use, or what you can get from places like Sportlogiq and Stathletes. There's also a definition issue, in that it's possible NST had a different location for high danger chances than HR (inner slot vs slot maybe?). Then you should consider context, as not all high danger chances are the same. On some the player is alone (breakaway), other times the shot is screened, etc., etc. Sample size of course needs to be considered with a stat like HDC, as the SV% can change quickly in a smaller sample. Therefore, you really should have an idea when data points stabilize and can be used with some confidence (believe it or not, to his credit Cam always asked this question on early XPG data).

Expected goals on ice and expected individual goals are both influenced fairly heavily by a player's OZ starts, which need to be considered when assessing a player. An aside, but it would be kind of cool to test the different XPG models that are publicly available to see which most accurately predicts XPG/XPGA/XPG% vs actual goals on ice. I'm not up to date on public hockey data; I know Natural Stat Trick, Hockey Reference, and NHL.com all have XPG data. Is there anyone else? Am I the only geek who would be interested in seeing how the different public XPG data compares?

I’ll read anything you write. Won’t understand most of it, but that’s not new.

 

[Over Guapo Grande]

I’ll read anything you write. Won’t understand most of it, but that’s not new.

Ditto. Though , I may get a bit more. I have like a 38th grade reading level.

 

[Myt1]

Ditto. Though , I may get a bit more. I have like a 38th grade reading level.

 

[Frisbetarian]

i would be interested. The public models are also built somewhat differently, which is why EH and dom can have variance in their evaluation based on the same raw data. I believe EH doesn’t use chance data for the reasons you stated with the public data but instead regresses raw shot data to account for game situation. Dom has said that he uses chance data. The nuance is why when I’m using those models evaluations, I look at all of them. It’s rare that there are huge variations between them so if all of them say player x is performing at y level, it feels reasonable to accept that as indicative of the players performance.

I think regular hockey fans are probably going to live in the gray until player tracking is easily available, the samples that creates are large enough to trust, and the models have been tweaked to push out accurate evaluation. What we have today is worlds better than we had ten years ago but there’s still a ways to go.

I’ll read anything you write. Won’t understand most of it, but that’s not new.

Ditto. Though , I may get a bit more. I have like a 38th grade reading level.

If I have some time over the next week I'll take a look at how well the 3 publicly available expected goal models correlate with goals on a player level over the past 3 seasons (5 on 5 only). Besides Natural Stat Trick, Hockey Reference, and NHL.com, are there any other available XPG models that I'm unaware of? I'll also take a look at how the public stuff compares to a few proprietary systems, but I unfortunately won't be able to say much about that here. But the first thing I'm going to do, inspired by @burstnbloom, is compare the 3 public XPG models to each other. I suspect, as he said, there is going to be very little difference between them.

@Myt1 and @Over Guapo Grande - If people as intelligent as you pendejos don't understand what I write, that's 100% on me! My job was to be able to explain analytical concepts so anyone could understand it, and I was OK at it.

 

[Smiling Joe Hesketh]

I'll be thrilled to read anything you write of this topic too, Chuck, this is fascinating stuff.

 

[cshea]

Ditto.

 

[Jordu]

I’ll be reading every word you write. Probably twice. Thank you.

 

[mwonow]

I’ll be reading every word you write. Probably twice. Thank you.

Me too - maybe more than twice, as my measured reading level never gusts higher than 34th grade. Ditto with the thanks as well!

 

[NYCSox]

Canes win but lose Raanta. They are on the verge of clinching division but may be down to third goalie for playoffs. WC1 is looking like best path possible for Bruins at this point.

 

[cshea]

This weekends results probably put us at like 90% to draw Carolina.

 

[durandal1707]

Looking at scenarios here...

.	Points	RW	ROW	Games
M1. CAR	112	45	51	@NYR	NJD
M2. NYR	108	43	47	CAR	MTL	WAS

Essentially the Rangers would have to win out and the Canes would have to lose their remaining two for New York to claim the Metro #1 seed.

.	Points	RW	ROW	Games
A3. TBL	106	37	47	CBJ	@CBJ	@NYI
W1. BOS	103	38	47	FLA	BUF	@TOR

Still possible to grab the 3rd seed in the Atlantic, but the Lightning have a much easier set of three games in terms of competition. However, the tiebreakers favor the Bruins currently.

.	Points	RW	ROW	Games
M3. PIT	101	36	42	EDM	CBJ
W1. BOS	103	38	47	FLA	BUF	@TOR
W2. WAS	100	35	39	NYI	@NYI	@NYR

Looks like a good chance that the Caps trade places with the Pens - only a point behind with a game in hand. The B's yet again have the worst schedule to close out the season here, but the good news is that the other two can only tie them in Regulation Wins, and the Bruins will have the ROW tiebreaker over both of them. So, if I have this right, any two points for the Bruins will secure the first Wild Card: Pittsburgh maxes out at 105 (and would fall to WC2 due to ROW), and if Washington maxes out at 106 they are guaranteed the 3rd Metro seed.

 

[cshea]

We'll see how they handle the final 3 games. My guess is playoff lineup with each of the goalies on Tuesday and Thursday. The Saturday game will probably be as much of a taxi squad game as they can muster (they can only call up 4 guys max from Providence).

There's no surprises on the playoff lineup. The forwards will be:

Marchand - Bergeron - DeBrusk
Hall - Haula - Pastrnak
Frederic - Coyle - Smith
Foligno - Nosek - Lazar

Lindholm - McAvoy
Grzelyck - Carlo
Forbort - Clifton/Reilly

My impression is that Clifton is the leader for the final D spot.

Ullmark
Swayman

I think Ullmark is the leader to start game 1. Butch has been pretty clear that they could rotate. He did say Ullmark would likely play Tuesday so that would have him in net for the two "rehearsal" games recently, Saturday vs. NYR and tomorrow vs. FLA. If Ullmark is indeed the game 1 starter, I think he probably plays on Saturday too to stay active given he just had a 9-day break.

 

[Frisbetarian]

I had some time and finally took a look at the different publicly available expected goal models to see how well they correlate with actual goals scored. As far as I can tell, and I could definitely be wrong about this, the only free and available expected goal data is on Natural Stat Trick and Hockey Reference. An aside, but if any of you cabrones know of another publicly available XPG model I’ve missed, let me know and I’ll run the numbers. I have also included Fenwick, Corsi, and a proprietary tracking system (not the Bruins), to see how they compared.

This work included data for the last 3 seasons (2018-19 through 2020-21), using all NHL position players with over 500 minutes in a season in 5 on 5 situations. This gave me a sample of just over 1500 player seasons. I ran a simple linear regression looking at the relationships between expected goals and actual goals, expected goals against and actual goals against, and expected goal percentage vs actual goals scored percentage when a player is on the ice. The results below are the coefficient of determination (r^2), which gives the proportion of the variation in the dependent variable that is predictable from the independent variable, or the proportion of actual goals/goals against/goal percentage which is predictable from the expected goal models.

Interestingly, but unsurprisingly, expected goals scored has a stronger correlation to actual goals scored in all systems than expected goals against to goals against. If you think about it, this makes sense because of the huge disparity in talent level among NHL goalies.

As @burstnbloom suggested, there is an extremely strong relationship between the Natural Stat Trick and Hockey Reference XPG data (r^2 G/GA/G% is 78.4% /72.5% /82.1%). Both Natural Stat Trick's and Hockey Reference's expected goal models track goals better than either Fenwick or Corsi. I was pretty surprised to see Hockey Reference's expected goals have a stronger relationship to actual goals than Natural Stat Trick, but it did both for each individual season, as well as for the combined data. The proprietary model I used was much stronger than either of the publicly available systems, which you would expect. This helps show how much better the data the analytically advanced teams have is (far more actions tracked and far more sophisticated).

Below is a table with the coefficient of determination for each relationship discussed above. I need to include a disclaimer here, as hockey has the highest degree of luck of any of the major sports, which, considering it's played with an odd shaped object and sticks on ice, makes perfect sense (if you're interested in seeing the numbers on this, let me know). So my confidence in the below is medium.

Model	XPG vs G	XPGA vs GA	XPG% vs G%
CF	26.7%	12.5%	21.8%
FF	29.6%	17.8%	24.1%
NST	33.8%	24.0%	28.7%
HR	40.1%	26.3%	32.4%
XX	49.8%	34.5%	44.1%

If you have any questions, let me know.

 

[durandal1707]

Great stuff as always, @Frisbetarian.

Moneypuck appears to compute their own formulation of expected goals as well. What they say about their methodology:

Shot Prediction Expected Goals Model

This model predicts the probability of each shot being a goal. Factors such as the distance from the net, angle of the shot, type of shot, and what happened before the shot are key factors in the model. This model was built on over 50,000 goals and 800,000 shots in NHL regular season and playoff games from the 2007-2008 to 2014-2015 season with location data. By adding up all the probabilities of a team's shots during a game, we can calculate the team's expected goals in that game. The model was built using gradient boosting. MoneyPuck's expected goals model uses a different variable strategy than other expected goals like from Corsica Hockey or HockeyGraphs.com. The MoneyPuck expected goals model does not explicitly use variables for rebounds or rush shots. Rather, it looks at the 'speed' between events: The distance on the ice between the shot and the event before it divided by the amount of time that's elapsed. Also, for rebound shots the model looks at the change in angle between the shots divided by the amount of time between the two shots.

 

[kenneycb]

How granular do you get with "actions"? Some are certainly easy to measure (completed pass vs. deflected pass vs. flubbed passes, etc.) but imagine it's near impossible to do others that are more fluid movements (winger went wide and took defenseman with him to open up the shooting lane).

 

[burstnbloom]

Awesome work, @Frisbetarian. It's good to see some correlation between the models. Shayna Goldman wrote a great article on the athletic comparing the methodology of the "core four" publicly available models (NST, Moneypuck, Evolving-Hockey and HockeyViz). Dom's model informs alot of their content but while his methodology is available, the data they use is not. I'd love to see how these all perform vs your proprietary model.

https://theathletic.com/3006028/2021/12/09/comparing-public-expected-goal-models-how-they-work-and-what-we-should-take-away-from-them/

 

[wilked]

View: https://twitter.com/_tyanderson/status/1519242083151843333?s=21&t=IkYel8Rl7Gmz7kNPdOKCqg

 

[TFP]

View: https://twitter.com/Tucker_TnL/status/1519351044865769474?s=20&t=6y-emwvShqmXrmrT7iKynQ

 

[FL4WL3SS]

Cool, with nothing to show for it.

 

[lexrageorge]

Cool, with nothing to show for it.

Finals appearance means a lot to some fans (myself included). And some really great playoff hockey that fans of other teams would kill for.

 

[jezza1918]

Cool, with nothing to show for it.

If you make it all about the destination for sure...but the team has been relevant and IMO fun to watch each regular season, and won at least one round in the playoffs each complete season (they lost in 1st round year he took over for Julien midseason).
It's been fun to be a fan of this team under his watch. Winning championships is extremely difficult and the older I get the more I seem to lean on the experience of the journey. That's not to say I won't be calling for his head when Forbert has a -3 in their final playoff loss this year.

edit: should be noted I totally respect those who are more of the 'title or bust' variety, if my post came off like I was talking down or something to those who don't view sports through the same lens I do then I apologize.

 

[cshea]

They've won a playoff series in each of his full seasons as head coach. That's not the ultimate goal, but it's a notable achievement. I believe they are the only team that has won a series each of the last 5 years.

It's really fucking hard to win a Stanley Cup.

 

[Salem's Lot]

I’ve criticized Cassidy numerous times over the last 5 years, but they made the finals with him behind the bench, that counts for something. They’ve only been to the finals five times in my lifetime, and I’m not a kid. Making the finals is an achievement for a coach.

 

[The Napkin]

Related, happy anniversary

View: https://www.youtube.com/watch?v=GQIL7HbsNF8

 

[FL4WL3SS]

Sorry, my comment was taken out of context. I meant it as more of "it sucks he doesn't have something to show for it". Meaning he deserved a cup.

 

[JohnnyTheBone]

I meant it as more of "it sucks he doesn't have something to show for it". Meaning he deserved a cup.

Fvckin' right! Game 5 vs St. Louis. Never forget.

 

[4 6 3 DP]

Sorry if this is a little game-thready but I went down the Youtube rabbit hole of watching most of the game 7s and other amazing moments of the run of this team back to 2011 (though soft place in my heart for Canadiens game 6 2008) and much like I felt in 2018 during that last Brady SB run, I just want these guys to get one more, I don't know if it's a Cup, or just one of those moments that the Garden roof blows off, but without getting too sentimental this group has created as many amazing memories as the second Brady run, as any of the Red Sox runs since 2003, and I really hope the offensive depth, the goaltending, whatever it needs to be can hold up to create a couple more of those moments for 37 and 63 particularly.

Anyways, no analytics here, just hope very much they get at least one more real run, whatever happens.