The TSN broadcast in Bruins vs Flames brought up an interesting side note about Berra's save percentage by period. It was interesting and I was wondering if it was the Flames or Berra who struggled late.
As a note, I didn't create this to slag on Berra. But the numbers posted on TSN are interesting and I was trying to figure out if it was systemic or individually (goalie) based.
Here's a raw data compilation. MacDonald was added for more data:
Spoiler!
Karri Ramo:
Period 1: 87 shots, 9 goals
Period 2: 132 shots, 15 goals
Period 3: 100 shots, 11 goals
Overtime: 3 shots, 0 goals
Reto Berra:
Period 1: 136 shots, 10 goals
Period 2: 126 shots, 9 goals
Period 3: 113 shots, 20 goals
Overtime: 11 shots, 2 goals
Joey MacDonald:
Period 1: 67 shots, 9 goals
Period 2: 59 shots, 4 goals
Period 3: 64 shots, 9 goals
Overtime: 66.67% on 3 shots
Here are the results: Save Percentage by period
Spoiler!
Karri Ramo:
Period 1: 89.66% on 87 shots
Period 2: 88.64% on 132 shots
Period 3: 89.00% on 100 shots
Overtime: 100% on 3 shots
Reto Berra:
Period 1: 92.65% on 136 shots
Period 2: 92.86% on 126 shots
Period 3: 82.30% on 113 shots
Overtime: 81.82% on 11 shots
Joey MacDonald:
Period 1: 86.57% on 67 shots
Period 2: 93.22% on 59 shots
Period 3: 85.94% on 64 shots
Overtime: 66.67% on 3 shots
So the obvious note is that Berra is the superior goaltender at the moment outside the 3rd. And it's not really like it's a systematic issue for the Flames: Ramo and MacDonald post typical numbers (for themselves) in the 3rd.
I'm not sure we can call them entire team collapses either. The shots surrendered in the 3rd are not atypical compared to the other period counts.
It could be that the Flames are not surrendering more shots, but are surrendering better shots in the 3rd. Unfortunately, I can't get a hold of a "shot quality" measure at the moment to compare what Berra and Ramo face in the 3rd. If someone interested could do that, that would be an amazing read.
The other outlier is MacDonald's performance in the 2nd. That appears to be a count thing though: Letting in even one more drops the save % by ~2%.
Curious what others read into these numbers though.
__________________
The Following 10 Users Say Thank You to kirant For This Useful Post:
In all fairness Berra has played 13 NHL games and Ramo has played 12 in the last 5 years. Not exactly a huge sample size. Normally goalies like these 2 would be brought in gradually as a backup behind a proven starter.
Also, if you look at the late goals against Berra you can usually see the rest of the team floating around watching the puck rather than playing their man. Growing pains.
The Following User Says Thank You to Jacks For This Useful Post:
In all fairness Berra has played 13 NHL games and Ramo has played 12 in the last 5 years. Not exactly a huge sample size. Normally goalies like these 2 would be brought in gradually as a backup behind a proven starter.
I would agree. The number of games is too small I think. It would be something best done after the halfway point (over 25 games each). But I had the spreadsheets typed out after the TSN comments, so I put them up just for interesting reading.
Quote:
Originally Posted by Jacks
Also, if you look at the late goals against Berra you can usually see the rest of the team floating around watching the puck rather than playing their man. Growing pains.
That too. But that also asks why exactly the Flames do that. Is it that they are just too comfortable to Berra playing amazing from the start of the game that they'll expect him to grab the last couple good opportunities? Or does Berra generate his own excitement? Another "easy to find numbers can't answer" thing...
__________________
The Following User Says Thank You to kirant For This Useful Post:
The sample size would make the stat fairly useless, but I'd be curious to see how Berra's save percentage in the last 10 minutes and the last 5 minutes of games compares to the other guys. My impression from watching the games is that it would be even more drastically worse.
Quote:
Originally Posted by nightfx
The obvious solution is to put Ramo in for the 3rd period and beyond
Obviously we care more about developing a future goalie than winning games, so it wouldn't make sense for us, but I'm always curious why no team has ever tried something like that. Maybe not so much for the 3rd period, but I still think Kipper should have had a designated shootout replacement for the past several years.
.823 SV% in the 3rd period is beyond awful and goes to show that some of the Flames 3rd period woes can be attributed to his penchant for giving up soft goals at the worst times.
The obvious solution is to put Ramo in for the 3rd period and beyond
And put Mike Morrison in for the shootout, obviously.
__________________
"For thousands of years humans were oppressed - as some of us still are - by the notion that the universe is a marionette whose strings are pulled by a god or gods, unseen and inscrutable." - Carl Sagan Freedom consonant with responsibility.
The Following 2 Users Say Thank You to evman150 For This Useful Post:
well, if you take into account that 5 of those goals alone come in 2 games in which the Flames had the lead against very good teams and tried to sit on it with Berra things look a little better.
Both are trying to adapt to the NHL and while its somewhat frustrating this year is going to be full of growing pains and we should come to expect that.
__________________
Thanks to Halifax Drunk for the sweet Avatar
well, if you take into account that 5 of those goals alone come in 2 games in which the Flames had the lead against very good teams and tried to sit on it with Berra things look a little better.
Both are trying to adapt to the NHL and while its somewhat frustrating this year is going to be full of growing pains and we should come to expect that.
This. You have to know that elite teams will push back being down late. That's what makes them elite teams, they can flip a switch and control the play. Chicago, Kane pretty much took over in the third. Boston, the whole team stepped up and turned it on.
Yeah a couple of the goals were soft, but at the same time Calgary needs to learn to kind of weather the storm that you know is coming and push back. I don't think Calgary sits on the lead, I think it's more of the other team building momentum and Calgary not really having an answer for it... it's like quick sand, they get caught, and they struggle to make the right plays, they get stuck, and before they know it they are in over their heads... winning is a learned process, and you learn it through failure. That is if you choose to learn from your failures rather then give up.
The Following 2 Users Say Thank You to TjRhythmic For This Useful Post:
i think earlier on, berra really let in some pure garbage goals in the 3rd period, but looking back at his past few starts, even last night, his sucky numbers go hand in hand with the team in front of him not knowing how to play with a lead in the 3rd period when the other team ups the "try level".
A handful of Berra's goals have been complete garbage, but looking at last night, each of the goals he surrendered where on good scoring chances. Similarly, Hall's goal in the last game wasn't Berra's fault. Hall was left wide open on the back side.
If you look at the first two periods, Berra does look more composed statistically and based on the eye test. It's just that both he and the team have a hard time keeping it up through 60 minutes. That has quite a lot to do with the lack of skill on the team at the moment.
__________________ Fireside Chat - The #1 Flames Fan Podcast - FiresideChat.ca
Small sample size for sure, but that doesn't mean it is useless.
One problem with small sample sizes is that one outlying event can continue to have a significant effect on the numbers for an inordinate length of time.
One way to eliminate this and get a more valid and useful assessment going forward is to now take a separate sample.
So we have the 12 games as one sample, that's good. Now, start over with a new 12 games (or 10 or whatever). That way, the previous numbers don't continue to pull down the results. If we see the same trends in the next sample, then we can makes some much stronger conclusions. If not, then sample size probably was an issue.
The Following User Says Thank You to Enoch Root For This Useful Post:
Stats or not, i think they are pretty even to me. Both have strength and weaknesses. Berra can cover the net and he actually play well when he is challenging the shooter. He gets in trouble when he sits back. Also being aggressive can have some flaws too. He wanders around a lot and it gets him in trouble and gets out of position. Ramo I think is more steady but he tends to give out big rebounds. The last three games he played, he kind of eliminate those rebounds.
Both goalies are still learning the NHL system and I think so far they are doing well. Berra is more like a bad luck that he can't win on a regulation time...giving up untimely bad goal..his fault or not. But he usually plays well on OT or shootout. Weird, huh?
Ramo vs Berra, by period
