This article is over 8 years old

What if Braden Holtby weren’t playing goalie for one of the worst defensive teams of the last decade?

All cards on the table: I believe Braden Holtby is the best goalie in Caps history. Presidents’ Trophies, Vezina Trophy, elite playoff performance (notwithstanding last year), you know the drill.

That said, it’s obvious Holtby is having a down year. I’ve said in the past that Holtby’s struggles in 2017-18 are the cost he bears for the team’s decisions last offseason, but it’s hard to be certain about stuff like that. Hockey is a complicated game with lots of moving parts, so I can understand why assigning blame personally on the goalie is popular with some people, who all seem to be Facebook commenters for some reason:

Ryan And don’t y’all dare blame Holts!!! Hahaha totally being sarcastic! It’s all him guys. We aren’t the best in D or even in goals but to make excuses for a goalie that is so “great” is just a joke. Bench his ass and watch what we do!!!

Mike Cant wait to hear all the its not Holtbys fault blame the defense. Caps held them to 19 shots on net Holtby should have saved them all. Fyi 0 shutouts this season. Not the same Holtby in net this season.

Michael Grubauer should be the starting goalie. Holtby is absolute garbage this season. Almost March and not one shutout, almost 3 goals against average, and 10 goals against in last two games. He is done. Trade is sorry ass and move on.

Eric If Holts hadn’t let in those 3 soft goals in the first, it’d be a different game. Caps completely dominated from the 2nd period on.

Robert Yep. Love holtz but he needs to sit for a bit. Not all in him but Grubs stops atkeast two of those.

(I did not edit the typos because I’m an unkind person.)

I actually think these people are generally right, but maybe for the wrong reasons. Yes, Holtby is not the goalie he was last season. Even he would agree to that. But I think the cause for the drop-off is both apparent and out of his control. If Holtby were backstopping another team right now, I think he’d probably be doing just fine. Not great, but fine.

Let’s first stipulate that Holtby is less than mediocre this season. Here’s a handy visualization brought to you by Micah Blake McCurdy of Hockeyviz.com:

This is a stark decline for Holtby, who has regularly sported a .930 save percentage during 5-on-5 and has perennially been a top goalie. The big difference from past seasons to this one seems to be how many opponent shots are coming from up close, as illustrated by another pair of visualizations from Hockeyviz.

I think it’s obvious which is which. The 2017-18 Capitals subject their goalies to more high-danger, point-blank, gut-shot chances than any team in the league – actually more than any team in the last decade, aside from two particularly horrible Coyotes clubs.

Team/Season Opponent HD per hour
ARI 17 13.3
ARI 16 13.3
WSH 18 13.3
NYI 18 13.2
NYR 18 12.8
NYI 11 12.7
CAR 11 12.7
CAR 13 12.7
ANA 18 12.5
CHI 18 12.3

Of the 37 goals Holtby has surrendered during 5-on-5 since the new year, 29 have been classified as high-danger according to Natural Stat Trick. Just one has been classified as low-danger. That suggests to me that Holtby is doing fine on more than 75 percent of this workload (the average and below-average shots) but is getting wrecked on the most dangerous shots, which are more numerous than they’ve ever been.

Those high-danger shots have really ticked up in February (15.0 per hour), an escalation that Holtby’s backup has not seen (11.8 per hour).

These numbers are based on Natural Stat Trick’s shot quality tiers. Corsica’s model disagrees with the implication that Holtby’s workload is easier than Grubauer’s. It expects Holtby to save .918 of shots (actual: .919) and Grubauer to save .910 (actual: .939). In either case, both Washington goalies are performing better (one slightly, one massively) than we should expect of them given the defensive troubles of the team before them.

And that got me thinking: What if Braden Holtby faced the quality of shots that other top goalies in the league faced? What if, instead of getting peppered with 13.1 high-danger shots per hour, he saw just 10.8 like Cam Talbot or 8.8 like Carey Price? Here is that hypothetical, with the What If column representing Holtby’s save percentages applied to that goalie’s workload.

Goalie Chance Save % What If Save % Diff
Anderson (OTT) 0.946 0.958 0.012
Price (MTL) 0.950 0.962 0.012
Talbot (EDM) 0.951 0.956 0.005
Lehner (BUF) 0.953 0.958 0.005
Lundqvist (NYR) 0.950 0.955 0.005
Dubnyk (MIN) 0.958 0.962 0.004
Hellebuyck (WPG) 0.957 0.960 0.003
Andersen (TOR) 0.958 0.958 0.000
Holtby (WSH) 0.953
Halak (NYI) 0.955 0.953 -0.002
Smith (CGY) 0.961 0.959 -0.003
Quick (L.A) 0.959 0.956 -0.003
Vasilevskiy (TBL) 0.963 0.959 -0.004
Gibson (ANA) 0.960 0.955 -0.004
Rinne (NSH) 0.963 0.956 -0.007
Bobrovsky (CBJ) 0.962 0.955 -0.007
Grubauer (WSH) 0.965 0.952 -0.012

This all assumes that Holtby would save the same percentage of shots at each quality tier, but would face different numbers of shots within each tier. Holtby out-performs seven of the 16 goalies here, ties one (Frederik Andersen), and is bested by eight, including Grubauer (who is included for comparison despite a much smaller sample). That’s still a mediocre performance from Holtby, but it’s a far better one than he is having on his own.

There is a very large caveat to an exercise like this: binning. By classifying all shots into just three categories (high-, medium-, and low-danger), we lose resolution along the margins and extremes. Lots of analysts consider over-reliance on binning to be naughty for that reason, and I don’t think they are wrong. This is more of an abstract thought experiment rather than a proper study, and I mean it only to illustrate that Holtby’s workload is what has driven him to the bottom of the pack in save percentage.

Also, just for curiosity’s sake, Corsica and Natural Stat Trick use different reckonings to classify high-danger chances, though the conclusions are similar. Here are save percentages per goalie against what each site considers high danger.

Goalie Corsica NSS
Grubauer (WSH) 0.861 0.912
Bishop (DAL) 0.856 0.910
Crawford (CHI) 0.859 0.896
Rinne (NSH) 0.851 0.907
Vasilevskiy (TBL) 0.845 0.907
Lehtonen (DAL) 0.846 0.906
Bobrovsky (CBJ) 0.854 0.895
Lundqvist (NYR) 0.821 0.892
Dubnyk (MIN) 0.812 0.901
Khudobin (BOS) 0.816 0.890
Elliott (PHI) 0.819 0.888
Rask (BOS) 0.818 0.886
Smith (CGY) 0.805 0.892
Quick (LAK) 0.821 0.879
Allen (STL) 0.820 0.875
Raanta (ARI) 0.807 0.886
Ward (CAR) 0.800 0.890
Markstrom (VAN) 0.805 0.881
Saros (NSH) 0.798 0.889
Jones (SJS) 0.820 0.869
Darling (CAR) 0.795 0.886
Mrazek (DET/PHI) 0.798 0.882
Condon (OTT) 0.793 0.885
Gibson (ANA) 0.803 0.874
Bernier (COL) 0.807 0.869
Nilsson (VAN) 0.809 0.867
Andersen (TOR) 0.802 0.872
Halak (NYI) 0.799 0.875
Hellebuyck (WPG) 0.774 0.883
Schneider (NJD) 0.810 0.855
Wedgewood (ARI) 0.804 0.866
Murray (PIT) 0.778 0.878
Talbot (EDM) 0.799 0.864
Fleury (VGK) 0.781 0.869
Hutton (STL) 0.769 0.871
Lehner (BUF) 0.779 0.860
Varlamov (COL) 0.755 0.870
Howard (DET) 0.773 0.860
Kinkaid (NJD) 0.789 0.843
Anderson (OTT) 0.774 0.852
Holtby (WSH) 0.768 0.859
Stalock (MIN) 0.764 0.849
Forsberg (CHI) 0.768 0.844
Reimer (FLA) 0.770 0.838
Dell (SJS) 0.746 0.844
Price (MTL) 0.755 0.838
Greiss (NYI) 0.690 0.828
Johnson (BUF) 0.695 0.805

The differences don’t change the rankings significantly. Holtby is ranked 41st out of these 48 goalies by Corsica’s reckoning and 38th by Natural Stat Trick’s. (Grubauer is ranked first in both.)

What’s curious to me is that Holtby’s lowly saving against high-danger shots doesn’t seem to be exclusive to him. His opponents seem to be struggling similarly.

Holtby Aggregate Opponent
High-danger shots faced 440 341
High-danger goals against 62 45
High-danger save % 0.859 0.868

That’s only a difference of one additional save on every hundred high-danger shots faced, and both of them are south of the league average, .875. I’m beginning to wonder if there’s something about how the Caps are playing this year – low-volume offense in fits and starts, allowing the opponent lots of on-the-rush and odd-man attacks – that is making high-danger chances at both ends of the ice more dangerous than other shots within Natural Stat Trick’s high-danger bin.

The 2017-18 Capitals are on the frontier here, and it wouldn’t surprise me if their systemic and nearly unprecedented failure to contain their opponents is hurting Braden Holtby more than we might expect. Either way, I don’t worry for Holtby in the long term. His team will get better, and maybe they already have. Holtby is probably still one of the best goalies in the league; we just can’t see it right now.

This article wouldn’t be possible without CorsicaNatural Stat Trick, and Hockeyviz. You can join us in supporting each of them on Patreon.

Headline photo: Cara Bahniuk

RMNB is not associated with the Washington Capitals; Monumental Sports, the NHLPA, the NHL, or its properties. Not even a little bit.

All original content on russianmachineneverbreaks.com is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International – unless otherwise stated or superseded by another license. You are free to share, copy, and remix this content so long as it is attributed, done for noncommercial purposes, and done so under a license similar to this one.

zamboni logo