# Success vs. attendance

This one should be obvious. If a team is winning, what happens to their home attendance? Goes up, right? In general, yes. But how much?

I was having a conversation with someone about attendance at lacrosse games, and he said that attendance had dropped at games in Philadelphia ever since the league started cracking down on hitting and fighting. It certainly hasn’t been eliminated from the game, but many think it’s down from where it used to be. He said that this is a bad thing for the league and this could be seen by looking at the attendance numbers. I pointed out that the fact that Philadelphia has had a playoff team only twice in the last decade may have something to do with declining attendance, so it’s pretty close to impossible to say that the drop in attendance was due entirely (or even partially) to the drop in hitting.

Hitting is something we don’t have accurate stats on, so we can’t really do any kind of analysis on how that correlates with attendance. But we do have won-loss records and attendance numbers, so let’s look at those.

What we’re looking for is how a team’s attendance correlates with that team’s success on the floor. To measure attendance (and factor out the number of games per season), we’ll use the average attendance at home games. To measure success, we’ll use the winning percentage, number of wins divided by number of games played. In this case, we are ignoring playoff games. I then calculated what’s called the correlation coefficient for each team. I won’t describe the math since if you know what it is you don’t need the description, and if you don’t know what it is you likely don’t care. Suffice it to say that a value of 1 means the attendance always goes up as success goes up and drops when the team is less successful. A value of -1 means it’s exactly backwards – attendance goes up as success goes down and vice versa. The closer the number is to 1 or -1, the stronger the effect – a value of 0 means that attendance and success are unrelated.

To avoid small sample sizes, we’ll only look at teams with 10 or more seasons in the NLL. The teams involved are the New York Saints, Baltimore Thunder, Philadelphia Wings, Colorado Mammoth, Calgary Roughnecks, Toronto Rock, Rochester Knighthawks, and Buffalo Bandits.

What this tells us is that the New York Saints attendance numbers were very dependent on their success – as their win-loss records started to decline, their attendance dropped. This effect was similar in Philadelphia, Rochester, and Colorado. The rest of the teams had much smaller coefficients, meaning that their attendance didn’t depend very much on their success on the floor.

Calgary’s value was negative, implying that as Calgary’s numbers go up, their attendance numbers actually go down. But this is a bit misleading – especially since I tweeted about it saying that it was depressing. The actual value is –0.019, which is close enough to zero that it’s fair to say that Calgary’s success on the floor is unrelated to their attendance numbers. The numbers for Toronto and Baltimore are slightly higher but still low enough to imply no correlation, and Buffalo is right at the bottom end of “low correlation”.

The definition of “bandwagon jumpers” or “fairweather fans” would be those who show up to support their team when they’re doing well and abandon the team when they’re not. Would it be unfair to refer to the numbers for the top four as being indicative of this? I’ll leave that determination as an exercise for the reader.

# Upcoming NLL milestones

Here are some statistical milestones that may or may not be reached during the 2013 NLL season. Of course, all of these numbers are completely arbitrary – does anyone discount Tom Marechek’s achievements because he “only” scored 399 goals in his career and didn’t get the elusive 400th? Of course not, but people seem to like nice round numbers, so here are a few that we may see this coming season.

## Players

A few players are close to some significant targets this season. Nobody will hit 1000 points unless a new single-season points record is set – Gavin Prout needs 119 points to reach 1000, and Dan Dawson needs 129. The closest points milestone would be 800, which only ten NLL players have ever reached, and which Mike Accursi will reach with only 15 more points.

John Tavares has reached more than his share of milestones over the years, and 2013 may feature yet another height to which nobody else has climbed. JT is already the only player in NLL history with 700 goals, and with 35 this season, he would reach 800. Tavares had 41 in 2012 so this is not out of the realm of possibility. To put this milestone in perspective, John Grant is the only other active player with over 500 goals, and Junior would need six more 50-goal seasons to reach 800.

Colin Doyle could join Tavares and Grant in the 500-goal club by scoring 42. Doyle has only reached 42 goals in one season three times and not since 2006 so this is a bit of a long shot.

Potential candidates for the 400-goal club include Lewis Ratcliff (needs 20), and Josh Sanderson and Mike Accursi, each of whom needs 32. Tracey Kelusky needs 43, but seeing as he’s only scored 32 in the past two seasons combined, this is even more of a long shot than Doyle.

A few almost-sure things are the three likely new members of the 300-goal club: Aaron Wilson needs 3, Blaine Manning needs 4, and Gavin Prout needs 7.

Just as John Tavares is the only player to reach 700 goals, he is also the only player to have reached 800 assists, and Tavares needs just 38 more to reach 900. Colin Doyle would need an 80-assist season (which has only been done once, right Mr. Billings?) just to reach 800. But only four players have ever reached 600 assists, and both Gavin Prout and John Grant are likely to join that club this season. Prout only needs 12 to get there, while Grant needs 26.

In terms of loose balls, no previously unobtained milestones will be reached this season. Or the next. Or even the one after that. Jim Veltman’s record of 2417 is safe for quite some time – John Tavares is the only player within eight hundred of Veltman, and Tavares would have to play for 3 1/2 more seasons (at his career average of 97 LB) to get there. But if anyone can reach Veltman’s astronomical total, Brodie Merrill and Geoff Snider are the most likely candidates, and each of them could reach the 1500 loose ball total this season. Merrill only needs 129 (he’s never had fewer than 157), and Snider needs 208, a figure he’s only failed to reach twice in six seasons. But assuming Merrill and Snider keep up their current paces (12.4 per game for Merrill, 14.5 for Snider) and play 16 games a year, Geoff Snider will become the new all-time leader 13 games into the 2017 season. I wrote about this once before and stated it wouldn’t be until 2018, but Snider’s 232 LB in only 14 games last year increased his average.

The 1000 loose ball mark is reachable by a few players: Gavin Prout only needs 27, Josh Sanderson 57, and Bill Greer 76.

## Teams

The Colorado Mammoth have three upcoming team milestones, two of which are very obtainable while the third will be close. The easy ones first: the Mammoth are 43 goals away from 2000 regular season goals, and they are 80111 people away from a total of 1.5 million in attendance, including both regular season and playoff games. An average of just 10013 per game in their 8 home games will attain that mark – and the Mammoth’s smallest crowd ever was 12537. The slightly more difficult milestone: 10 regular season wins will give them 100.

The Philadelphia Wings’ first loss of the 2013 season will be their 150th regular season loss in their history. This is far and away the most losses for any NLL team, and nobody else is even close. Of course, they’ve played at least five more seasons than anyone else. But consider this: in the last ten seasons, the Wings have only reached .500 three times (and only exceeded it once). The fact that they are still above .500 all-time is a testament to how good they were in the 1990’s. There were only 8-12 games per season, but the Wings had seven straight seasons over .700. In fact, regardless of how they finish this season, they will still end up above .500. Even going 0-16 this season will put them at 169-165.

The Calgary Roughnecks’ first home game will be their 100th, and the Riggers could reach one million in regular season attendance this season as well. They are 66896 away from that mark, an average of 8362 per game. The ‘Necks averaged 8313 per game in 2012, so just an extra 50 people per game will do it.

The Edmonton Rush should have a much easier time reaching their attendance milestone than the Roughnecks. The Rush only need 19201 to reach the half-million mark. Other than a slight bump from their first season (2006) to their second, the Rush’s average attendance has dropped every season. But unless it drops by over 30% from 2012 to 2013, they should hit the half-million target in game 3.

## League

This is a fact that I first pointed out on my personal blog back in 2008, and then reposted on The NLL Blog in 2010 (and have since seen mentioned elsewhere as well): The last time the NLL began a season with the same teams in the same cities as the previous season was 1993. Barring last-minute foldings like the Ravens in 2005, 2013 will end the 19-year streak. If you’re looking for stability in a league that’s shown anything but for almost two decades, this might be the biggest milestone of all.

# Goals per game in the NLL

When hearing someone describe the NLL to a non-lacrosse person, you tend to hear the same things over and over:

• played in a hockey rink with the ice covered with artificial turf
• similar rules to hockey, but with the shot clock and over-and-back rules of basketball
• high-scoring, average of about 25 goals per game

But how accurate is that “25 goals per game” number? On the surface, it seems about right – games like 14-10 or 13-12 are pretty typical, 18-15 is a little on the high side, and 11-7 is a little low. But if we actually crunch the numbers, what do we find?

Amazingly, we find that this number is almost exactly correct. Taking into account the 1,633 games (regular season and playoffs) from 1987 up to and including the 2012 season, the average number of goals scored per game is 24.99. But the breakdown by season is surprising:

The first ten years or so were pretty unpredictable, ranging from 22.6 in 1990 to 29.1 only two years later. The extremes: the highest scoring season was 1992, when 29.1 goals were scored per game. 2011 was the lowest scoring season, with an average of only 21.7 goals per game. The first six seasons were interesting – two seasons in the mid 27’s, two low-scoring seasons of 24 and 22, then the two highest ever, 28.2 and 29.1.

The obvious trend is that from 2000 until 2011, the number of goals scored dropped pretty steadily, from 28.2 in 2000 to a low of 21.7 in 2011. The NLL increased the width of the nets from 4’6″ to 4’9″ in 2002, and one of the first games of the 2002 season featured the Montreal Express defeating the Calgary Roughnecks 32-17. Fans wondered if that would be the norm with the new nets, but in the end it made little difference; the average actually dropped about half a goal from 2001 to 2002, and then down over a full goal the next year as goalies adapted. However in 2012, a number of rule changes were made in an attempt to speed up the game, and seemed to have the (possibly unintentional) effect of increasing scoring as well. After the lowest-scoring season ever in 2011, scoring rebounded in 2012, jumping 2½ goals per game to 24.2.

Why did the rule changes increase goal scoring? Here’s why:

• The 8 second rule (instead of 10), the “immediately drop the ball on possession changes” rule, and the fast starts all meant that there were more transition chances, and many of those were converted.
• In addition, the faster the transition, the more likely that an offensive player will get stuck on the floor playing D, and some offensive players are just not the two-way players of old. They’re not all as skilled at their own end of the floor as at they are the other end, and so playing five top offensive players against four defenders and one O guy playing D gives the offense a bit of an advantage.
• Defenders were also forced to give up their longer 46″ sticks for 42″ sticks, obviously making it harder for them to stop the John Grants and Dan Dawsons of the league.
• Finally, on a five-minute power play, three goals are now required to allow the penalized player out of the box instead of two. I don’t think this rule came into effect all that often, but it did mean that some 5-on-4’s lasted longer in 2012 than they would have in 2011.

# Will the NHL lockout affect the NLL?

When it became clear that an NHL lockout in 2012-2013 was inevitable, many NLL fans, writers, players, and executives seemed to believe that while this is a big drag for hockey fans (which many lacrosse fans are), it could be a good thing for the NLL. The obvious logic is that without hockey to watch, hockey fans may look for other places to spend their sports event dollars. What better place to spend it than on a league that plays in many of the same arenas, with a similar sport, featuring some of the best athletes in the world, and with tickets that cost a fraction of NHL tickets? The NLL can’t lose! Can it?

Of course, this has happened before. The NHL missed the entire 2004-2005 season due to a lockout, and so the entire 2005 NLL season was played while there were no NHL games being played. How did the NLL do attendance-wise that year? Let’s have a look.

The overall average attendance for the NLL in 2005 was 10237, which was up 3.6% from 9885 in 2004. Looks promising so far. But 2004’s attendance was up 14.3% from 8649 in 2003, so attendance was already increasing. In 2006 the attendance was 10703, which was up 4.6% over 2005. Overall attendance did increase in 2005, but less than it had in 2004, and less than it would in 2006.

Here is a graph showing the average home attendance for each team as well as for the entire league (the black line in the middle).

Do you see any peaks in 2005? Toronto has a little one, but they won the Championship with a powerhouse team. 2005 was the last year of the Rock’s early-2000’s dynasty so the increase makes sense. Buffalo was right in the middle of their impressive resurrection from only 7002 in 2003 to the mid-15000’s only 4 years later. Arizona was up 13.9%, but that’s all of 800 people. Other teams showed no significant increase, if any. Calgary was up 2.1%, but grew 15.4% the next year. Anaheim, San Jose and Toronto were up less than 2%, and Rochester less than 1%. Colorado was down 3% and Philly was down almost 14%.

The obvious but unfortunate conclusion is that the 2004-2005 NHL lockout had little to no impact on NLL attendance.

Unfortunately, I don’t have any TV numbers, so I can’t look at whether more people watched the NLL on TV in 2005. If the NLL approached TSN or Sportsnet or the CBC (Anyone for “Lacrosse Night in Canada”? Could it happen?) with the opportunity to televise NLL games, that could be a ton of exposure for the league. I don’t pretend to have the faintest idea on how the finances of such a deal would work; it could be that the NLL would have to pay for that privilege rather than receive money from the networks. I know the Rock paid to have some of their games televised over the last couple of seasons.

Could the NLL benefit from the NHL lockout? Attendance-wise, it doesn’t seem likely without the league doing a fair bit of work (and possibly spending a fair bit of money) to advertise the hell out of the league and bring new people in. After that, it’s up to the league to continue that push to make sure all those first-timer’s come back once hockey starts again.

# Defense wins championships – or does it?

Offense wins games. Defense wins championships.

I’ve heard that quote in relation to lacrosse, football, and basketball, and it’s probably been applied in other sports as well. I know that a lot of Toronto Rock fans from the early 2000’s believed it, but is it generally true? Let’s take a look at the NLL Champions from 1987 to 2011, covering 25 seasons. Don’t worry, this isn’t nearly the propeller-head stats-fest that my article on back-to-back games was.

I went through each Championship team and calculated their rank in the league that year in terms of both goals scored and goals against. Just so we’re clear, “first” in goals scored is the highest amount, while “first” in goals against is the lowest. That’s fairly basic and obvious stuff, but I wanted to spell it out to avoid any misunderstandings. I’m going to ignore the absolute value of goals for and against, mainly because a different number of games were played in different seasons. Using the rank rather than the value factors that out, as well as other differences like rule changes. I’ll look at goal scoring trends in the NLL in a future article.

The “rank” I’m using for a given team is “1 plus the number of teams that are ahead of the team in question”. So if two teams scored more goals than the team I’m looking at, they are ranked third. If another team scored the same number of goals, then the team I’m looking at was actually tied for third, but I’m ignoring that – tied or not, they still had the third-highest total.

Before we get to the general trends, here are the extremes. In the 25 years of the NLL, only once has the Championship winner been both #1 in goals scored and #1 in goals against – the 1994 Philadelphia Wings. At the other end of the spectrum, the 2003 Rock were ninth in goals scored (they scored 36 fewer goals than the #1 Bandits that year), but first in goals against. The 2007 Knighthawks were the exact opposite – first in goals scored (scoring 30 more goals than anyone else) but ninth in goals against.

If defense wins championships, then it stands to reason that most Championship teams would rank higher in goals against than they would in goals scored. But we don’t find that to be the case. Out of 25 seasons, 12 of the Champions (or 48%) ranked first in the league in goals scored, but only 8 (32%) ranked first in goals against. The average rank for goals scored is 2.6 while the average rank for goals against is 3.1. This means that on average, the Championship team is closer to the top of the league in goals scored than they are in goals against, i.e. most Championship teams are better offensively than they are defensively. Defense does not win championships.

But there was a period where it did. From 1998 to 2003, the Rock won four titles and the Wings won two. Only one of those teams – the 2001 Wings – was not first in the league in goals against, and only those same Wings were as high as third in goals scored. The Rock Championship teams in 1999-2000 and 2002-2003 were 5th, 6th, 7th, and 9th respectively in goals scored. But before that period, the top defensive team had only won the Championship once, the 1994 Wings, and it’s only happened once since, the 2009 Roughnecks. On the other hand, from 1988 to 1996, every Champion except one (the 1990 Wings) was first in goals scored. It didn’t happen again until the 2005 Rock, but then it happened in five of the next seven years.

Here’s a graph of the ranks of the Championship teams in both goals scored (blue) and goals against (red). Notice how the blue line stays low until about 1997, then jumps up for a few years before dropping back down again. At the same time, the red line is higher during the 90’s, then drops down to the bottom while the blue line is high, then grows again when the blue one drops. That inversion was the Les Bartley era in Toronto.

Generally, the NLL Champions have been better offensively than defensively. But as we’ve seen, from about 1998 to 2003, that trend was reversed. Of those six seasons, the Toronto Rock under Les Bartley won four Championships – and lost a fifth to the Wings in a low-scoring defensive game. This is one reason Bartley was so well-respected – not only because he led the Bandits to the only undefeated season in NLL history, but because he bucked the trend and built a team that was a defensive powerhouse rather than offensive, and was exceptionally successful doing it.

This is not to say that you don’t need a good defense to win, of course you do. And it’s not to say that you can’t win with a great defense and adequate offense. It’s just happened far more often in the past the other way around.

# Could happen…

As of now, three teams have already clinched playoff berths: Colorado, Calgary, and Philadelphia. But given the parity in the NLL this season, it’s far from settled who else will make it. Here are a few scenarios that could still happen:

### Bandits win the East

Buffalo is having one of the worst seasons in its history but unbelievably, they could still win the East Division. If the Bandits win their four remaining games, they end up at 8-8. They already own the tiebreaker with the Rock and if they win out, they’ll own the Knighthawks one as well. As long as the Rock win no more than twice and the Knighthawks don’t win out, the Bandits finish no worse than second. The Wings own the tiebreaker with the Bandits so if they win even once more, the Bandits can’t catch them. But if they lose out, the Bandits win the east outright. Not bad for a pathetic stupid team with no heart.

### Knighthawks win the East

If the Knighthawks win out and Philly loses twice, they end up tied at 9-7, with Rochester owning the tiebreaker. As long as the Rock don’t win three times, the Knighthawk win the East.

### Rock miss the playoffs

If the Rock lose their last four, they end up 6-10. As long as Buffalo and Rochester each win twice, the Rock finish last in the East. If Minnesota beats Philly twice, Edmonton beats Toronto twice and Calgary once, and Washington beats Minnesota, Toronto, and Buffalo, they all finish at 7-9 and the Rock are out.

### Philly finishes last in the East

If Philly loses out, they end up 7-9. If the Rock beat Edmonton twice, they have 8 wins. Rochester will get a win against Philly and if they beat Calgary twice, they’ll also have 8 wins. If Buffalo wins out, they finish with 8 wins too, and the Wings are last. As I said, the Wings have already clinched the playoffs; in this scenario, Edmonton will lose three more games, putting them at 6-10.

### The Wild Wild West

Calgary and Colorado have locked up first and second in the West – each has 10 wins and nobody else can end up with more than 9. But I think there are scenarios where each of Edmonton, Minnesota, and Washington can come in third, fourth, or fifth, and in some of those cases, fifth place will cross over and make the playoffs.

# Are back-to-back games a disadvantage in the NLL?

In a recent article on IL Indoor, Teddy Jenner examined the teams that have played back-to-back games this season. He discovered that more than two thirds of the teams that had two games in a weekend won the second of those games. The results were better for teams playing at home on the second night and less so if playing away.

Those are pretty interesting numbers, but they only cover three-quarters of one season. If only we had such statistics on previous seasons – but for that we’d need information on all previous NLL games. But wait! We have that! Let’s fire up Graeme’s Super-Amazing Magic NLL Statistical Database-inator™!

I did some queries looking for two games involving one team played within 3 days of each other. This will include not only games on consecutive days, but also games played on Friday and Sunday of the same weekend. I ended up doing four queries: the team in question plays at home both games; away both games; home first and then away; and away first and then home. I then combined these numbers for the aggregate record. We’ll deal with home-and-home series (i.e. both games involved the same two teams) below.

### The Numbers

From 1987 to 2011, I found 394 instances where a team played more than one game in a weekend. Here are the numbers:

Type Games Win-Win Win-Loss Loss-Win Loss-Loss
Home-Home 5 1 1 2 1
Home-Away 166 44 44 33 45
Away-Home 148 45 26 38 39
Away-Away 75 12 18 20 25
Totals 394 102 89 93 110
Totals (%)   25.9% 22.6% 23.6% 27.9%

Strangely, teams have played two home games in the same weekend only 5 times, but teams have played two away games 75 times. The percentage totals show that when a team played two games in a weekend, the most common scenario is that they lost both games. Winning the first and losing the second is the least common.

Looking at the numbers a different way:

Type Wins first game Wins second game
Home-Home 2 3
Home-Away 88 77
Away-Home 71 83
Away-Away 30 32
Totals 191 (48.5%) 195 (49.5)
Home Totals 90 (52.6%) 86 (56.2%)
Away Totals 101 (45.3%) 109 (45.2%)

So 48.5% of the time, the team won the first of back-to-back games, and 49.5% of the time, they won the second. Unsurprisingly (?), the numbers are better at home.

But the real question is not “what was the most common outcome of such series?” but “can we make inferences or predictions based on past behaviour?” To answer that question, we must do some statistical analysis.

### Statistical analysis

Here’s where we get into the stats stuff a little more. Don’t worry if you’re not a stats geek, I’m not going to describe the tests I did in great detail or describe how or why they work, primarily because I have no idea. (Or more accurately, I don’t remember since it’s been over twenty years since I studied this stuff.)

Our default assumption, or “null hypothesis”, is that each of the four outcomes of a two-game series (win-win, win-loss, loss-win, loss-loss) is equally likely. Basically, we make no assumptions that there are any patterns of any kind, and let the numbers tell us if we’re wrong. Our observed values are shown in the chart above, so we want to see whether the differences between those values and the values that our null hypothesis would predict are statistically significant.

As an analogy, say we flipped a coin 100 times. Our null hypothesis is that each outcome is equally likely (i.e. our coin is fair), so we’d expect 50 heads and 50 tails. If our actual results were 52 and 48, that’s likely to be close enough for us to conclude that our null hypothesis is probably correct. If our results were 70-30, we’d reject our null hypothesis and decide that one outcome is more likely than the other, and so we likely (the numbers can’t conclusively prove anything) have a rigged coin. But what if the numbers were 58-42? Is that close enough to 50-50 for the differences to be insignificant, or is it more likely that your coin is unfair?

We can use a test called the chi-squared test to calculate the probability that the differences between a group of observed results and the expected results are due solely to chance, or if it’s more likely that there’s something else involved causing the differences. I calculated (well, Microsoft Excel calculated) this probability using the chi-squared test, though I omitted the Home-Home row because all of the expected values were too low. (Chi-squared doesn’t work very well for expected values below 5.) The p-value calculated was 0.059793, or about 5.98%. What this means is that the probability that the values we observed would have been observed if our null hypothesis was true is almost 6%. To be considered statistically significant enough to reject our null hypothesis, this value should be less than 5%.

The long and the short of is it that we cannot reject our null hypothesis. The numbers do not indicate that playing two games in a weekend has an effect on the likelihood of winning either one. Playing two games in a weekend is no different than playing two games a week apart.

### Home-and-home

Now let’s look at home-and-home series. Note that there has never been a weekend where two teams played each other twice in the same location, so we’re only dealing with each team playing one game at home and one away. There are four possibilities here: a split where the home team wins both games, a split where the away team wins both games, a sweep where the home team wins the first game, and a sweep where the away team wins the first game. There have been 53 such series’ in NLL history (1987-2011):

Type Games
Sweep – Home, Away 16
Sweep – Away, Home 16
Split – Home wins 12
Split – Away wins 9
Totals 53

The most common occurrences have been sweeps, but after applying the same chi-squared test as above, I came up with the p-value number of 0.453534, or about 45.4%. This is way over the 5% required to be statistically significant so these numbers really tell us nothing, likely because of the small sample size. Similarly, the numbers do not indicate that any outcome of a home-and-home series is more likely than any other.

### Conclusions

To summarize the conclusions I’ve drawn above:

• The evidence does not indicate that playing two games in the same weekend affects the likelihood of winning either game.
• The evidence does not indicate that any of the four possibilities of a home-and-home series is more likely than any other.

A team playing two games in a weekend is no different than that team playing two games a week apart. The numbers tell us that it simply doesn’t matter. There are always going to be outliers, but for every team that loses the second game because they’re tired, there’s another team that’s energized from playing the night before.

Just to be clear, these numbers don’t tell us that there is no pattern. They simply say that the data does not indicate a pattern. It also tells us that we can’t use the numbers to make predictions; in the past when a team played two games in a weekend, the most common outcome was that they’d lose both games. This does not mean that in the future when a team plays two games in a weekend, they are more likely to lose both than any other outcome.

Many thanks to Dan Shirley from In Lax We Trust (and a math undergrad at Washington State University – go Cougars!) for his help in interpreting the statistics.