2015-2016 Trends

2015-2016 Capital Trends Game 19

So the Caps are just about a quarter of the way through their season and we’ve seen some good and some less than good in both the results and the process. I know the win last night may make most Caps fans happy, but there’s been a few troubling trends over the last several games. I think the Caps can be better than they have been. Of course I’m only basing that assumption off of the games earlier in the season; but I guess we should start to take a look at what’s working and what’s not working.


Playoff Pacing


Not a whole lot of streaking in any particular direction here. However, they have had a casual improvement in their pace since I last looked at their performance (at game 11 – so 8 games ago). This is good on the whole, since it’s not really like they’ve had the benefit of any particular PDO benders, unlike the Rangers who seem to be in a perpetual PDO bender. Thanks Hank.



SAT_bars_Game_19 USAT_bars_Game_19 YTD_possession_game_19 rolling_possession_game_19

Game-by-game, there’s been a very mixed bag in the possession department. Their games against Detroit were notably bad, as was their one game against Boston. Over the last 8 games though, the possession is falling in every metric. It’s a very stark pattern in the 10-game rolling data. They seem to be better when looking at SAT vs USAT in most visuals. I’m curious if there’s much less of a focus on shot blocking, given their current defensive corps. This might be a pattern to keep an eye on with Orpik still out. Here’s another thought I had on the Orpik injury:


YTD_shots_game_19 YTD_ES_shots_game_19 rolling_shots_game_19 rolling_ES_shots_game_19

The Caps had the best shot suppression figures in the league. They still are one of the top teams, but they’re number 2 now, bested by Carolina (wut?). The silver lining is that at least they’re maintaining a very healthy shot differential. While the shots against spiked up, the shots for did as well. A team that manages ~5 more shots per game over their opponent should definitely find themselves in the playoffs. And that’s all we care about right now. Anything grander than that goal would be putting the cart before the horse right now.


PDO & Percentages

YTD_PDO_game_19 YTD_percentages_game_19 rolling_PDO_game_19 rolling_percentages_game_19

The Caps were looking pretty unfortunate over their 10 games leading up to the Dallas game, but since then, they’ve been the recipients of a few good bounces. I can recall a few instances (maybe only 1) where iron saved the Caps from relinquishing another goal. I still fully expect better from Holtby. There’s nothing wrong with what he’s posted so far, but he’s better than this. You don’t get your name into the elite goaltender discussion on accident. The offense for the Caps has me a little less optimistic. Their ratio of high danger scoring chances to shots on goal has seemed to dip. Even if they are hitting a lot of goalposts and having goals overturned, I’d like to see more legitimate scoring chances.


Special Teams: Summary

YTD_ST_game_19 rolling_ST_game_19

The special teams are looking not so special lately. And I think I know the culprit…


Special Teams: Power Play

 YTD_PP_Performance_game_19 YTD_PP_Shots_game_19 rolling_PP_Performance_game_19 rolling_PP_Shots_game_19

Yep. The Caps power play has not seemed very powerful. I had high hopes for this year’s variation with Oshie in the slot, but they’ve barely been sputtering along. They’re getting their chances and drawing enough penalties, but they’re not shooting very efficiently. I would like to see more shots in general, but that doesn’t seem to be fluctuating too much one way or the other, they’re just not finding the back of the net. Carlson and Ovi need to put in some reps on that feed. It’s just not working. Ovechkin with 1 power play goal in 19 games is not something I would have imagined in any scenario.


Special Teams: Penalty Kill

 YTD_SH_Performance_game_19 YTD_SH_Shots_game_19 rolling_SH_Performance_game_19 rolling_SH_Shots_game_19

There’s not much to dislike here. The Caps’ PK unit has been keeping shots down, and they’ve been staying out of the box. Even with the dip in save percentage in the rolling data (an artifact of the small sample size of PK opportunities – seriously look at that green line dive), the PK is definitely something I’ve been pleased with. Now if only they could keep this up, while righting the power play. Everything would be sunshine and roses.



This is still a very good hockey team. The power play should be fixable. Ovechkin will start to put some in the net while up a man. The team has managed to avoid some of their older troubling habits, and they haven’t been turtling when leading (granted – they haven’t been leading much in their recent games). However, they’ve allowed their slow start habit to persist. Hopefully, last night’s game is an omen of that pattern reversing.

It’s almost Thanksgiving, and the Caps are firmly in playoff position. That’s a good position to be in, historically speaking.

2015-2016 Trends

2015-2016 Capital Trends Game 11

Sorry it’s taken me 11 games to write up a post, but I didn’t want to have just noisy, turbulent data to present without some form of a pattern forming, so we at least have an idea of who this year’s Caps are.

A couple quick notes:

  • I’m trying to move away from the legacy names for the possession proxy measurements. I’m going to start using the names denoted by NHL.com’s wonderful enhanced stat site, simply because they’re more intuitive. So Corsi is now SAT or shot attempts, and Fenwick is now USAT or unblocked shot attempts. These will usually be expressed as percentages of the total, as I’ve done in the past. If expressed as a rate or raw number, it will be USATF for unblocked shot attempts for, or SATA for shot attempts against, instead of FF or CA.
  • I’ve also changed the algorithm for adjusting these shot attempt metrics. I’m using Micah Blake McCurdy’s (@IneffectiveMath) algorithm that he published. And he was generous enough to provide the coefficients he’s currently employing in all the visuals on his site. So these numbers now reflect a weighting based on score state and venue (home vs away). This provides a stronger correlation historically than any other commonly used methods.

Now onto the Caps…

The team is quite a bit different from last year’s version. But you already knew that. However, there doesn’t seem to be any form of an adjustment period for the system this time. That’s actually quite impressive given the amount of turnover this team saw over the summer. There are a couple individuals whom you’d expect to have some growing pains. Guys like Backstrom and Orpik who had offseason surgery, and therefore didn’t have any real practice. Then there’s Dmitry Orlov, who hadn’t played a professional game in like 18 months. Mix those elements with the uncertainty of how newcomers Oshie and Williams would fit in, and couple that with how the younger guys would progress, or how other players would be thrust into more responsibility, and it’s not difficult to place a lot of uncertainty with how this team would perform out of the gate. So we should all take that into consideration whenever we weigh in on the shortcomings we’ve witnessed so far, whether individual or teamwide.

I made that sound like there are a lot of shortcomings for this team. But there really aren’t, as we’ll see. But this is just to hopefully help balance any negative feelings we have still in our consciousness after that Rangers game. Here we go!

Playoff Pacing

The Caps are off to a strong start. I know, you don’t make the playoffs in October, but you can miss them. We can also say that the Caps have mostly only beaten mediocre teams, which may be mostly true. But, let’s also remember that in previous seasons, the Caps could usually be counted on to drop quite a few games to “inferior” opponents.



SAT_bars_Game_11 USAT_bars_Game_11


I’m going to mostly focus on the score adjusted SAT data. As this has become the norm in the analytics community for assessing teamwide performance now. So in that department, the Caps are looking good. They’re on the right side of 0 in all games but their first two, and they have a couple truly dominant performances. Those aren’t all wins, and wins are ultimately what matters, but this is an encouraging sign. They’re above 55% for their season cumulative SASAT (score adjusted shot attempts). That’s championship caliber territory.

PDO & Percentages

YTD_PDO_game_11 YTD_percentages_game_11

They were riding some pretty high PDO earlier, but that’s come back down. There’s a bit of mixed fortune in that though. Fortunately, we should expect Holtby to increase his save percentage drastically. He’s well below his historical averages right now. We have no reason to expect that to continue. I’m not saying it’s impossible, but his .905 even strength save percentage is notably lower than his .930 cumulative even strength save percentage coming into this season. I expect to see him bounce back. And to start making some more of those impressive saves we’re used to. His high danger save percentage is about 10 percentage points lower than what he usually achieves.

The other side of that equation is the shooting percentage though. The Caps are shooting at a pretty high level. I want to believe it’s sustainable. They have a lot of elite offensive talent. And they also seem to be a team that has a proportionately higher percentage of their events in the high danger category.
They operate more frequently in the…
Danger zone.

To prove this point, although I feel the sunglass emoji does that for me, let’s take a look at a chart from war-on-ice:


Here, we’re mostly looking at the y vs x axes. The x axis is how many shots on goal a team generates per 60 minutes. The y axis is the rate at which a team generates high danger scoring chances. The bubble color is shaded based on teamwide on-ice shooting percentage. If you imagine a linear regression line to this data, you’d expect teams that lie below the line to shoot with less efficiency (and be colored red), while teams above it would shoot with greater efficiency (and be colored blue). Washington is to the right and below Philly, and to the left and below Dallas. So they should be pretty clearly above the imaginary line. I’d expect San Jose to progress towards a blue bubble if they continue their current trends, same goes for Philly. Maybe Montreal is getting a little lucky, again. But for the Caps, I’d expect them to shoot above league average if they keep up their scoring chance generation. Whether, they continue to shoot so high above league average, is probably a little less likely.

Special Teams: Summary


Special Teams: Power Play

YTD_PP_Performance_game_11 YTD_PP_Shots_game_11

As you’d expect given their current skid, just about everything is coming down a bit for the Caps. The shot generation is only decreasing slightly, so that’s encouraging. Maybe they’ve just hit a bit of an unlucky spell, because they’re still generating shots well. Or maybe John Western Union Carlson needs to stop telegraphing his passes to Ovi so clearly. I know there’s been plenty of teeth gnashing over the Caps’ powerplay and a lot of analysis has been done based on who’s on the point, without any real definitive answer. But I may start to look at Ovi individually, and see if his shot generation is suffering on account of RAHJC.

Special Teams: Penalty Kill

YTD_SH_Performance_game_11 YTD_SH_Shots_game_11

Things are still a bit noisy here. Although, it’s comforting to see the frequency at which the Caps are heading to the penalty box is on the downswing. The shot suppression is a little concerning. But the shots getting through to the net are still below 2/minute, which is good.

That’s all I have for now. And I think that’s plenty. In future installments, I’ll look at individual score states. So we can see if the Caps have a killer instinct, or if they’re turtling, or any other fun metaphors. For the time being, the Caps are looking swell, and most of the data backs up the eye test. But as a word of caution: the Caps also looked really good in most departments last year too. So we’ll see if they can keep up the good work.