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.

Saturday, December 26, 2009

Scoring Chances for Game Number 20262 : Wild @ Capitals

Scoring Chances for NHL Game Number 20262

MIN118:10 8212236376724161740855v5
MIN113:29 361114253739212840895v5
WSH15:30 3692537 914192240524v5
WSH15:14 3692537 914192240524v5
WSH15:00 3692537 914192240524v5
WSH13:38 36142537 23172128404v5
WSH13:37 361425375523172128405v5
WSH13:29 3611142537310394053895v5
MIN10:09 3891114373394053 5v3
WSH217:03 5111437515539212840895v5
WSH214:58 6892137 917212840524v5
WSH214:21Goal522253755 914212840524v5
MIN213:41 361114375124141619405v5
WSH213:05 8111436375124103940535v5
MIN212:49 591520375539212840895v5
WSH23:50 3692037 314192840524v5
WSH23:07 81114253637316172140895v5
WSH20:57 68920376724141719405v5
WSH20:26 3620253767921262840525v5
MIN315:29 8212236376724163940535v5
MIN315:05 369152037316405385895v5
WSH314:10 511142537552492128405v5
WSH39:36 81520213637310141940895v5
MIN38:38 361114375124172140855v5
MIN36:39 31421363751310394053895v5
WSH35:36 589111537921262840525v5
WSH34:16 81114375155310394053895v5

3M. ZIDLICKY16:46533:16003:0005
5K. JOHNSSON19:52132:58001:1401
6G. ZANON20:11540:01003:4306
8B. BURNS16:27372:32001:4301
9M. KOIVU14:07222:30002:4905
11O. NOLAN16:24372:53000:3900
14M. HAVLAT16:53472:37000:1201
15A. BRUNETTE14:14222:08000:0000
20A. MIETTINEN14:12232:24002:1301
21K. BRODZIAK11:51420:01001:1401
22C. CLUTTERBUCK12:48300:01000:5101
24D. BOOGAARD5:30010:00000:0000
25E. BELANGER12:01152:31001:5605
36J. SCOTT8:56340:00000:0000
37J. HARDING49:079125:02004:5707
38R. EARL5:21010:00000:0000
51J. SHEPPARD12:35330:00000:0000
55N. SCHULTZ17:04141:17000:1401
67B. POULIOT12:53320:01000:0000

PeriodTotalsEVPP5v3 PPSH5v3 SH

Tuesday, December 22, 2009

Scoring Chances for Game Number 20257 : Wild @ Lightning

Scoring Chances for NHL Game Number 20257

TBL119:36 36915203245121330775v5
TBL118:06 81221223234811213038435v5
MIN116:19 361114253245121330775v5
TBL115:33 52432515567819212630915v5
MIN115:03 52432515567819212630915v5
MIN114:29 8915203255511303843775v5
MIN111:23 891520323429111417305v5
TBL10:13 3692132 45122630914v5
MIN216:51 3624325167517192830775v5
MIN216:13Goal359152032821303843 5v4
TBL214:15 59213255 45122630914v5
MIN211:52 3511142532811192130 5v4
MIN211:46 3511142532811192130 5v4
TBL211:17Goal811142532341214307791 5v4
MIN24:52 35915203225303843 5v4
TBL22:29 58122122322451326305v5
MIN21:17 3624325167813263077915v5
MIN316:17 35915203225303843 5v4
MIN314:57 511142532552453038 5v4
MIN314:45 3691520321430384377 5v4
MIN314:27 3612212232412142630775v5
TBL310:44 512212232345891130385v5
TBL38:49 89203255 45122630914v5
TBL37:35 36111432342591113305v5
TBL30:14Goal69203255 45121326914v5

3M. ZIDLICKY19:09524:21602:2101
5K. JOHNSSON19:48244:12603:2501
6G. ZANON20:55520:31104:2102
8B. BURNS17:38222:23010:5601
9M. KOIVU11:21213:23403:5904
11O. NOLAN16:24213:23310:5700
12C. KOBASEW8:54140:15000:0000
14M. HAVLAT16:42113:02310:0700
15A. BRUNETTE9:07213:14400:0000
20A. MIETTINEN11:19212:53403:0502
21K. BRODZIAK13:08140:21002:3702
22C. CLUTTERBUCK11:17140:15001:3200
24D. BOOGAARD10:18410:00000:0000
25E. BELANGER15:07102:47311:4900
32N. BACKSTROM51:13976:31717:0304
34S. HNIDY8:48140:29010:0700
51J. SHEPPARD10:45510:00000:0000
55N. SCHULTZ17:24311:06102:5603
67B. POULIOT12:19510:00000:0000

PeriodTotalsEVPP5v3 PPSH5v3 SH

Sunday, December 20, 2009

Even Strength Data for Games 1-500

As a continuation of a previous post, I figured that I'd throw up EV data for the first 500 games of this season.

I've included both overall EV data, as well as data on even strength play with the score tied. Empty net goals have been excluded.

The first sheet shows data for games 1-500, the second for games 1-252, and the third for games 253-500. I've also included a fourth worksheet that compares each team's corsi ratio with the scored tied from the 1st half of the year (games 1-252) to the 2nd half (253-500).

Some comments:

It pains me to admit it as a Habs fan, but Gabe Desjardins is absolutely correct -- the Canadiens look like a terrible team by the numbers. Over the last 250 some games, the Habs have had the worst corsi ratio in the league at EV with the score tied, and by a large margin at that. The scoring chance numbers don't look any better.

To make matters worse, the Habs have also been getting bombed on special teams. In the same 250 game period, they've given up almost twice as many shots on the PK as they've accumulated on the powerplay (55 vs. 98). In fairness to Montreal, that has more to do with their league worst penalty differential than it does with special teams performance per se. In any event, the numbers aren't good.

Curiously, the team's underlying numbers were actually quite respectable over the first 250 some games. It'll be interesting to see where they end up.

On the other side of things, the Ducks appear to have improved considerably at EV relative to the first 250 games. While their current record may not be impressive, they're definitely trending in the right direction. If they can find a way to take fewer penalties, they should be able to at least compete for a playoff spot.

Phoenix continues to perform well at EV. The season is almost halfway over at this point and the Coyotes currently have the 2nd best corsi ratio in the league when the score is tied. This is surprising considering that they were 28th last year in this regard.

I'm not entirely sure on how to account for their turnaround, although I suspect that it boils down to two things. For one, they've gotten rid of and/or sent to the minors a lot of guys that were really hurting them last year (Turris, Lindstrom, Hale, Fedoruk, Lisin, Porter, and Carcillo). The replacements -- Lombardi, Prucha, Fiddler, Lang, Vandermeer, Aucoin, and Vrbata -- are demonstrably better hockey players.

Secondly, I think that the coaching change has likely had an effect as well. Tippett's teams in Dallas were consistently able to outshoot the other guys, both at EV and overall. While it's hard, if not impossible, to quantify his contribution, I think that it's safe to say that he's an upgrade over a relatively inexperienced Gretzky.

Scoring Chances for Game Number 20502 : Wild @ Canadiens

Scoring Chances for NHL Game Number 20502

MIN118:45 31214213255622313674915v5
MIN114:51 59202432341526314074755v5
MTL111:59 57142132482631323675915v5
MIN110:46 3691520321315202231465v5
MIN18:51 91222323455622313246915v5
MIN15:52Goal59141520322631747591 5v4
MIN13:14 920243234552031323647915v5
MIN12:58 5791415321520314047745v5
MIN12:17 3621223248613142231465v5
MIN217:46 512222532551520314047745v5
MTL216:10 2432343851551720314767945v5
MTL214:43 5791432481314263140755v5
MTL210:27 369152032613142231465v5
MTL210:24 369152032613142231465v5
MTL29:58 369143248622313236915v5
MTL28:39 57122532511420314674755v5
MIN21:12 59152032551326314675915v5
MTL313:07 61432344851613142231465v5
MIN39:56 5714324851613142231465v5
MIN37:59 3614324851613142231465v5
MTL37:16 3691520322631327475915v5
MTL33:12 5712222532613143246475v5
MTL33:11 5712222532613143246475v5
MIN32:38 369152032613143246475v5
MTL30:32 3692132 1314464774914v5
MTL30:03 3692532 1314464774914v5

3M. ZIDLICKY19:37550:43002:1302
5K. JOHNSSON20:43651:15100:5800
6G. ZANON21:24560:00002:1302
7C. STONER14:33250:00000:0900
9M. KOIVU18:54751:15102:1402
12C. KOBASEW15:13440:43000:1400
14M. HAVLAT15:04441:58100:0900
15A. BRUNETTE16:33431:15100:0000
20A. MIETTINEN16:51531:15100:1400
21K. BRODZIAK11:25210:00001:2401
22C. CLUTTERBUCK15:35430:00000:5200
24D. BOOGAARD7:44310:00000:0000
25E. BELANGER12:57240:43001:3301
32N. BACKSTROM54:4214121:58103:2002
34S. HNIDY15:39420:00000:0900
38R. EARL7:10110:00000:0000
48G. LATENDRESSE13:51340:43000:0000
51J. SHEPPARD12:49330:00000:0000
55N. SCHULTZ17:28610:00000:5800

PeriodTotalsEVPP5v3 PPSH5v3 SH

Olivier at En attendant les Nordiques has been recording scoring chances for the Canadiens this season. His chance count for the above game can be viewed here. Overall, our numbers are fairly similar.

Saturday, December 19, 2009

Scoring Chances for Game Number 20234 : Stars @ Wild

Scoring Chances for NHL Game Number 20234

MIN117:40 91520323455320283137915v5
MIN116:20 891520325526262931 5v4
MIN115:10 8122122325559142831965v5
MIN112:30 361417253259141728315v5
MIN110:15Goal922323455 36101431634v5
DAL19:32 1225323455 59212931914v5
MIN18:05 36112232 36103163964v5
DAL17:41 36112232 36103163964v5
MIN16:27 8914153234214172331375v5
MIN13:31 361520213223102631635v5
MIN12:06 369152032621293137915v5
MIN219:05 361417253225212931915v5
DAL216:34 369152032517203137965v5
DAL216:27Goal821253255 59212931914v5
DAL213:46 9152032343623212931915v5
MIN212:19 3614172532514233137965v5
DAL211:41 81124325567610262831635v5
DAL210:39 6122122323423212931915v5
DAL28:02 8122122325523212931915v5
MIN26:53 81114325567610172831635v5
MIN25:25 81221223255621282931915v5
MIN25:24 81221223255621282931915v5
DAL24:10Goal620253234 510213163914v5
DAL20:53 36212532 2392131 4v4
MIN319:31 39142232 3263137 4v3
MIN319:24 39142232 3263137 4v3
DAL312:59 611213255 56103163 4v4
MIN312:09 389141532321233137 5v4
MIN311:56 3691415322691731915v5
DAL39:54 81124325567610262831635v5
MIN37:21 3811142532310313763 5v4
MIN36:03 361415253226293191 5v4
MIN35:51 361221223226213191 5v4
DAL33:46 36202532 59212931914v5
DAL32:57 891221325536101431635v5

3M. ZIDLICKY14:49825:57603:1912
6G. ZANON18:37840:35204:4013
8B. BURNS15:03544:31301:2701
9M. KOIVU13:31535:34401:3310
11O. NOLAN12:10230:54102:0211
12C. KOBASEW9:24430:12101:3901
14M. HAVLAT12:48605:43500:2400
15A. BRUNETTE11:16624:20300:0000
17P. SYKORA9:45301:59000:0000
20A. MIETTINEN10:56422:29102:4402
21K. BRODZIAK12:07550:12103:1601
22C. CLUTTERBUCK9:38321:30301:1821
24D. BOOGAARD6:25020:24000:0000
25E. BELANGER9:55312:43203:1004
32N. BACKSTROM43:091498:48708:0325
34S. HNIDY13:27222:28003:1712
36J. SCOTT8:59010:00000:2400
55N. SCHULTZ15:29552:41102:5912
67B. POULIOT9:38120:24000:0000

PeriodTotalsEVPP5v3 PPSH5v3 SH