Thursday, December 31, 2009

Shot Recording Bias: Florida and New Jersey

Earlier this year, I made a post that examined whether certain NHL arenas systematically undercount or overcount shots on goal. My methodology involved comparing each team's seasonal home and road splits from 1995 until 2008 in terms of shots on goal . More particularly, the total number of shots taken by both teams in a given team's road games was compared to the total number of shots taken by both teams in that same team's home games.

Where the home-road split revealed a discrepancy in recorded shots, I then looked at the shooting percentage data to determine whether there was, in fact, a recording bias. I reasoned that if a discrepancy was due to bias, rather than randomness or other factors, there ought to a concomitant increase (in the case of undercounting) or decrease (in the case of overcounting) in the shooting percentage of both teams in games played in the arena in question.

Since that time, others, such as Sunny Mehta, Vic Ferrari, Chris Boersma and Tom Awad, have also examined the issue through work of their own, all of which I would recommend reading.

Two of the arenas that I suspected might be overcounting shots were Bank Atlantic Center in Sunrise, the home of the Panthers, as well as Continental Airlines Arena (and, perhaps, the Prudential Center as well) in New Jersey. While the data on shooting percentage suggested that shots were likely undercounted in New Jersey, the same was not true of Florida.

The purpose of this post is to take a somewhat more refined look at the topic so as to properly determine the existence of bias. While my initial post looked at overall shot totals and overall shooting percentage, it failed to consider the influence of specific game states, such as special teams play and the playing to the score effect. As both of these factors are known to influence shots on goal as well as shooting and save percentage, merely examining the data in aggregate can be more misleading than illuminating. In order to mitigate these concerns, the data below has been broken down according to game situation.

Firstly, the data for Florida. Presented below is a table comparing the number of shots taken by both teams in Florida road games versus Florida home games, broken down by game state and season. Shots that resulted in an empty net goal have been excluded. This essentially confirms what was already known -- that more shots are recorded in Sunrise than elsewhere. Interestingly, the putative bias appears to be confined to even strength, with no effect on special teams.


I've also prepared a similar table that compares the shooting percentage (again, for both teams) in Florida road and home games. As with the previous table, the percentages do not include empty net goals.


Looking at the data, it's difficult to argue for any sort of shot recording bias. The aggregate shooting percentage in Florida home games is identical to the aggregate shooting percentage in Florida road games. The same is essentially true at even strength with the score tied. If shots were, in fact, being overcounted, then one would expect to to observe a lower shooting percentage in Florida home games. But such is not the case.

In the comment thread of this post made by the Contrarian Goaltender, Vic Ferrari surmised that some of the apparent shot recording biases are not biases at all, and that some arenas really do consistently feature more or fewer shots than average, perhaps due to team style, strategy or some other like factor. I think that's probably the best explanation in this case. The Panthers, for whatever reason, seem to play a more exciting style of hockey at home, which serves to increase the shots on goal numbers while leaving the shooting percentage data unaffected. This is consistent with the shot discrepancy being restricted to even strength.

The data for New Jersey tells a different story. Unlike in Florida, New Jersey home games have featured a deficit of shots, rather than an excess.


More significantly, however, this deficit in shots has been accompanied by an increase in the shooting percentage in Devil home games. This implies that the deficit is due to recording bias, rather than some other factor.


Looking solely at even strength play, the shooting percentage in Devil road games from 2003-04 to 2008-09 has been nearly an entire percentage point lower than the shooting percentage in Devils home games during the same period. While the difference may not seem large, it is greater than what one would expect to observe through chance alone. I've included a separate table below that shows a range of expected shooting percentage values, expressed in the form of confidence intervals, for both EV play with the score tied as well as for EV play in general.


This table shows the range in values where one would expect the overall shooting percentage for New Jersey home games to be found, during the period under review (2003-04 to 2008-09), if it is assumed that:
  1. There is no shot recording bias
  2. The 'true' shooting percentage in Devils home games is equivalent to that observed in Devils road games.
As a specific example, consider the Devils home-road splits at EV. The Devils and their opponents had a combined EV shooting percentage of 0.072 in Devils road games played between 2003-04 and last season. Thus, it is assumed that the underlying shooting percentage in Devils home games is 0.072. Making the further assumption that shots are recorded accurately in New Jersey, the table shows the range in the 'expected' shooting percentage for Devils home games. So, for example, if the above assumptions are true, one would expect to see the shooting percentage in Devils home games fall between 0.067 and 0.0777 95% of the time, and between 0.0654 and 0.0795 99% of the time. The observed value was 0.081, which lies outside both confidence intervals.

One final comment: some will have noticed that more shots were recorded in New Jersey home games than road games for both 2007-08 and 2008-09. I take this to mean that the shot recording bias is likely no longer in existence. While it is true that the shooting percentage in Devils home games continues to be higher than in Devils road games, the difference is probably meaningless in the absence of an actual difference in recorded shots. Perhaps the switch to a different arena was accompanied by a change in shot recorders.

10 comments:

Sunny Mehta said...

Good stuff, J.

Even though there were more shots recorded at home in NJ in the last two seasons doesn't necessarily mean there wasn't bias. (I.e. it's possible that maybe even MORE shots should've been counted.)

The 08-09 home/road EV Sh% discrepancy is pretty crazy. If we use the Road number as our hypothetical mean, the home number is way out of range. But if some other number in between is our hypothetical mean, it's possible that chance made it so that we ended up with two numbers on opposite sides of the bell curve. (Of course another explanation is that the Devils have an ability to score goals at home better than other teams do - dubious.)

This leads me to want to see that type of chart for every team, and to know what league average EV Sh% was on the road and at home. It might give us a better idea of where to place the hypothetical mean.

Vic Ferrari said...

Terrific work, JLikens.

07/08 and 08/09 are years when Brent Sutter coached in New Jersey. He prefers a more uptempo game, as evidenced by his preference to go power vs power instead of using a checking line.

I mean if you prefer a slower pace of game, the opponent is more likely to oblige in your rink. I think that if we looked at games between VAN, STL and ANA vs the southeast division ... the games played in the east were generally more open and entertaining and featured more shots at net.

Does that make sense?

Sunny Mehta said...

Vic,

Agreed on the Sutter point. He was also preceded by Julien, so his changes look even more drastic. Couple questions.

1) Why do you think the opponent is more likely to oblige in the home team arena, and how much difference do you think it can hypothetically make?

2) Can you think of a reason why Sh% and Sv% numbers would be affected by that? A few months ago I noticed the huge gap in the Devils' home/road EV Sh% last season, but the thing is, they led the league in EV goals at home - hard to cry bias there. Do we have any reason to believe teams can be more efficient in goal scoring based on whether they're home or on the road?

Olivier said...

Vic, that's an interesting point and it leads me to ask a question around: remembering last year's comments on the arrival of Blysma with Pittsburgh, some commenter noted Therrien's teams had an habit of getting outshot. Are shots, as a metric, enough to tell us something of the abilities/style of a coach?

Vic Ferrari said...

Sunny:

Just generally, a lot of teams want to play a more entertaining style on home ice. Especially if the rink isn't sold out.

If you don't care about that, then the road team is more likely to oblige.

When VAN rolls into Nashville the GM, coach and ownership probably all want to see some high event hockey.

When NSH rolls into VAN, especially if they are tired, they're going to be happier to play the game 10 feet off the boards and keep things at a slower pace. And since that's the way Vigneault like's to play it ... probably what you'll see.

So I suspect that there still was undercounting of shots in N.J in 07/08 and 08/09. And I suspect that if you went through the H2H icetimes for a bunch of N.J games, you'd see that Parise played tougher minutes on home ice.

Just a theory, but it would mesh with Brent Sutter's style.

Vic Ferrari said...

Olivier,

I think there were a lot of things that changed around the coaching change. And probably Therrien wasn't a very good coach, either.

Mostly Gonchar coming back and Crosby playing healthy after a rough year, I think. The big change in underlying numbers came in where Malkin's shifts were ending. I guessed at the time that Bylsma may have "gotten through to him". But seeing some of the playoffs, Malkin got a whack of chances in the first 10 seconds of coming onto the ice after a Crosby shift. Having Sid healthy and able to play PvP would send a huge ripple down that lineup.

I've never checked to see if that was the case in the regular season games after Bylsma took over, but that's what I would guess. Just my sense of it.

JLikens said...

Sunny:

Yeah, I was thinking about which value I should use for the hypothetical mean.

While I ultimately elected to use the aggregate road SH% in Devils road games, those values look to be a little low -- almost a whole percentage point below the league average SH%. It's possible -- maybe even likely -- that the true shooting percentage in Devils home games is closer to the league average (both at EV with the score tied and EV generally).

Vic:

Interesting point about Sutter. I suppose the difference in coaching styles is capable of explaining why the percentage discrepancy persists even though the shot gap has closed.

It seems, then, that the bias continues to operate.

London Recording Studios said...

Resident started in 2001 by taking over the ailing Stepping Stone Studios, which was at one time a leading studio in the busy Willesden music scene and had a substantial studio complex, albeit deserted. London Recording Studios.

Hen Party London said...

Hen Party London experience can give you glamour, sophistication and the ability to replay your single or pop video at the actual wedding! This hen party experience will stay with all of you forever and will only be a DVD player a way.

Host Pay Per Head said...

Well it is something that you don't want to find. Discrepancies even though they aren't good but you can find interesting things.