The unbalanced 2017 North American Soccer League schedule and mid-week match attendance

January 28, 2017

There is a fair bit of uncertainty regarding the upcoming Jacksonville Armada season. Questions abound such as  “Where will the team play?”, (almost certainly not its home for its first two seasons, Community First Park, AKA the Baseball Grounds of Jacksonville), “Who will be in goal? (the team does not have any goal-keepers under contract at this time), and so on.

One thing that is certain is the nature of the season. On January 20, the North American Soccer League announced that 2017 league play will consist of eight teams will play sixteen games in both a spring season and a fall season.  After thinking about it for a moment, my reaction was similar to that of Florida Times-Union write Clayton Freeman.


Reaction on Twitter from supporters of NASL teams featured a bit of puzzlement, as well. While many were happy that the league’s previous practice of having a spring season that was half the length of the fall season (but whose titles counted equally) was a thing of the past, there was doubt over why the schedule wasn’t balanced. Why not just have 8 teams playing their opponents twice each season such that there would be two 14 game seasons? It seemed so odd to me that I almost seriously wondered if someone made a “fencepost error” and mis-calculated the number of games that would make up a balanced schedule.

The acting commissioner of the league, Rishi Sehgal, has answered the question in a consistent fashion during interviews. This, for example, is from Sehgal’s interview with Raleigh based writer Neil Morris

“Primarily because we wanted to give more games back to the fans. It was a long off-season without soccer, and the desire was to create more excitement about the competition rather than create a short season. I’ve read rumors out there that we had penciled in other teams to play and then canceled them. That speculation is just speculation. The decision was made to have as many games as possible and focus on the competition.”

There is something about adding “bonus” games in this manner that has bothered me. There are 32 weekends from the season-opening weekend to the weekend before the playoffs start (March 25-October 28).  In the past, and Sehgal suggests that this will be the case in 2017 as well, there is a break between the two seasons. That means that there aren’t enough on which to play games. Some games will be played mid-week.

As an Armada fan, this concerned me, because there was a noticeable difference between attendance for the weekend matches and weekday matches last season. I mentioned this to Clayton Freeman on Twitter and he helpfully replied with the stats:

Admittedly, the drop of 43% wasn’t quite as bad as it seemed. One could argue that the number attendance dropped from was at least as problematic as the drop itself. That is, dropping 43% from a weekend of average of 8,000 isn’t that bad. When you’re only getting 4,000 fans on the weekends, though, the result is games that are money-losers. (Playing in an active baseball stadium, the Armada paid a city subsidized, yet still hefty fee, for converting the baseball field to a soccer pitch; Empire of Soccer cited a cost of $70,00 per match)

Being curious as to whether such drops were common across the league, or a problem unique to Jacksonville, I compiled the numbers for the past two seasons across the league. The figures are pretty interesting.

In 2015, eleven teams played each-other once in the Spring and  twice in the Fall. That meant for five home games in the spring and ten in the fall.  During the fall, most teams played two home games mid-week, while Atlanta played three and Edmonton, Indy, and Tampa Bay played only one. The following table shows each team’s average weekend attendance, weekday attendance, and the percentage difference between the two.


Atlanta 4294 2947 -31%
Edmonton 2872 3120 8.6%
FTL 4858 1955 -59%
Indy 9956 8500 -14%
Jacksonville 8261 5760 -30%
Minnesota 884. 8275 -6%
NY 5314 3522 -33%
NC 4728 3307 -30%
Ottawa 5228 4751 -9%
San Antonio 7036 4509 -36%
TB 5619 4691 -16%


A few of these numbers are skewed a little bit. Fort Lauderdale, Jacksonville, and New York all had season openers with attendance that was out-sized compared to the rest of the season. Adjusting those towards normal attendance improves the drop by 5% or so for each team.  Un-adjusted, the league-wide figure suggests an average attendance drop of 25% from weekend to mid-week.

As one can see, not everyone was hurt by mid-week games; the 3,120 at Edmonton’s compared favorably to some weekend games that had 2,000 or fewer fans in attendance. Similarly, neither Ottawa’s or Minnesota’s games ranked atop the list of most lightly attended home matches. Elsewhere, however, the attendance drops were significant.

In 2016, the Spring season consisted of eleven teams (but not the same eleven as in 2015, since Miami and Oklahoma City replaced Atlanta and San Antonio) playing each-other once. Puerto Rico joined in the Fall season to make its season twelve teams playing each-other twice.  This made for five home games in the Spring and eleven in the Fall. Jacksonville played one home one-game mid-week in the spring; the only team to do so. Jacksonville played four mid-week games at home during the Fall; again this was exceptional as all other teams played only two home games mid-week. (One of those mid-week Jacksonville games was not scheduled as such; Hurricane Matthew pushed a would-be weekend match against Indy into the week.) As was the case in 2016, some teams, particularly the southern ones, took significant attendance hits during the week.

Edmonton 2092 1836 -12%
FTL 1484 437 -71%
Indy 8585 7075 -17%
Jacksonville 4055 2278 -44%
Miami 5259 4854 -8%
Minnesota 8775 7154 -18%
NYC 3829 3402 -11%
NC 5323 3201 -40%
OKC 3209 3472 8%
Ottawa 5591 4700 -16%
PR 4350 1329 -69%
TB 5992 5076 -15%

The figure for Fort Lauderdale is not a typo; they only had 873 fans total at their two mid-week matches. The crowd for Puerto Rico’s  mid-week games were roughly half the size of that for their most lightly attended weekend match.  As was the case in 2015, Jacksonville and Carolina had big drops in attendance for mid-week games.  The drop for the Cosmos wasn’t as bad… but the figure for weekend-games was worse than in 2015.  Edmonton,Indy, and Tampa managed to keep the gap reasonably small.  The figures for Miami and OKC are a bit wonky. Miami had some games with horrible attendance in the Spring.  OKC managed to have their mid-week games before the franchise started to fall apart. Otherwise, their numbers would have likely been comparable to Fort Lauderdale’s.

Pardon a slight digression as I elaborate on the Armada. As mentioned earlier, Jacksonville played in an active baseball stadium; their home schedule was dictated by the home schedule of the AA baseball team.  As home-stands that wrap around two weekends are not uncommon, the Armada were left with some tight windows for home games, hence the excess number of mid-week games.  Cruelly, those were the games most poorly attended. The crowd for the mid-week Spring game was ~500 fewer than the lowest attended weekend game. Of the four Fall games, only one had higher attendance than the lowest week-end game (that week-end game had 2,516 fans in attendance; it was played on a scorching afternoon in advance of a Jaguars pre-season game). Weather was not a factor for the depressed attendance. In 2015 the only games affected by weather were the season opener (15 minute downpour in the second half) and a Saturday night game that was delayed by lightning. Similarly in 2016, the only games with unpleasant weather were during the weekend. The only mid-week game that had understandable poor attendance was the rescheduled Indy Eleven match; there were still Jacksonville residents without power on the night of that match.

League-wide, I calculated the 2016 drop in attendance from weekend to weekday games to be 27%, nearly identical to the drop observed in 2015. The drop isn’t so big to suggest that the average fan does not go to mid-week games. It is certainly large enough, however, to suggest that adding games to the schedule yields diminishing returns if the games in question are going to be played mid-week. A 28 game schedule fits into the given March 25-October 28 window fairly nicely; 14 week-end games each season and each team gets a week off as well as a two week break between seasons. Fitting a 32 game schedule into that window with a a an equal amount of rest means that each team would have to host two mid-week games each season.

If franchises are going to be on the hook for that many mid-week games, it appears that some will have to put some serious focus on how to make those games better attended. It certainly appears that Tampa did something that the other southern teams didn’t.

The other puzzler with the unbalanced schedule is how it’s going to work. It would be more evident if the teams were more evenly distributed geographically and everyone had a local rival. Under those conditions, it would make sense for the extra games to be merely just an extra home and away game against the rivals. Alas, the 2017 NASL is not like that.  Here’s my best effort of making the unbalanced 32 game schedule work in a cost-efficient manner :

  • Each team plays each-other home and away twice
  • Puerto Rico, Miami, and Jacksonville play each-other home and away an extra time.
  • Indy, New York, and North Carolina do the same.
  • San Francisco and Edmonton: Y’all are going to love to hate each-other!  Two extra matches, home and away!

This, of course, could make things a wreck balance of competition-wise. From a competition stand-point, the only fair way to do it would be to assign the extra games pseudo-randomly.  As a few have noted though, that could bring delights such as Puerto Rico making extra trips to Edmonton and San Franciso, etc. Adding those extra trips, for any team, doesn’t make sense.

Furthermore, it doesn’t make sense to  have extra games that one can reasonably expect to have reduced attendance. As a season-ticket holder who is concerned about the feature of his team and its league, I am not in support of this unbalanced schedule as it has been presented.






Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: