The Making of the NWHL on NHL19

I spent hours making the 2019-2020 NWHL in NHL19. Here’s how that went...

I’ve had NHL19 more or less since it came out. I’ve played the Be A Pro mode (only as a woman, not that the game uses she/her pronouns anyway), played as the New York Rangers, and played as a created expansion franchise (Atlanta Mustangs, if you were curious). Before the NWHL even postponed the Cup, I had an inkling from the news cycle that they probably would need to, and devised a plan.

My plan was this: to give myself something to do other than think about how anxious I was, and to give other people the opportunity to watch sports at a time when there are essentially no live sports being played anywhere on the planet, I would try to make the Isobel Cup happen. Kind of.


I started with spreadsheets. Mine and Mike Murphy’s (he is a saint).

I made my own, with a page for each team. Listed all the players, grouped by position, on each team. Listed their names, their position, their age, their dominant hand, their hometown, and made a column for ‘player class.’ In addition to having an overall rating (OVR) out of 100, players in NHL19 also have ‘player classes’, which are as follows:

Forwards: Sniper, Two-Way Forward, Playmaker, Grinder, Enforcer, Power Forward

Defenders: Offensive Defenseman, Defensive Defenseman, Enforcer, Two-Way Defenseman

Mike also helped me fill out some of these player classes for certain teams. Shoutout to my homie Mike.

Each player also has a ‘potential.’ For example, a player might be a potential elite player, potential franchise player, or their potential might be simply a seventh defender. This matters less for what I was doing, but I still assigned them, because it does help the AI create lines when the teams play.

In order to estimate what each player’s OVR rating should be, I used statistics from the 2019-20 season instead of trying to rely on some kind of objective assessment of player talent. Doing that would be almost impossible. Probably completely impossible. There are definitely players in the NWHL whose statistics do not reflect their actual value to their teams or actual, objective talent level. I tried to reflect who I thought was underrated by giving them OVR ratings a bit higher than their statistics would have suggested.

Creating Players

First of all, there are an abysmally small number of face options for female players in NHL 19. I did the best I could, but frankly, most players basically look the same. In addition, you can’t make a player under 5’7’’, so most players are taller than they actually are in real life. I tried to maintain some kind of scale, ie: the shortest player on a team, if she’s 5’3’’, becomes 5’7’’, so everyone else is also 4 inches taller.

The hair options are also terrible. Why are there so many options for mullets? I’ve never seen a women’s player with a mullet. Why are there MULTIPLE mullet options and no long hair options other than high buns? Nobody wears buns like that under a helmet! Where are the long braids, the long ponytails? And why are all of the hair colors so awful? What do we need a white AND a silver for? Why is ‘faded brown’ kind of green? Why is the blonde so dark? What is the point of having red AND orange but no real, light blonde color?

Anyway...once I got past the point where I understood none, or very few, of these players would actually look like themselves, I moved onto the important stuff.

I learned very quickly that manually inputting ages and hometowns would be absolutely insane. It would have taken me forever and there was really no point because those things didn’t change the gameplay at all. Similarly, trawling through the play-by-play name options to find something even remotely close to most NWHL players’ names is not only time-consuming but mostly disappointing. I also realized that I had listed players’ ages but what I really needed was their birth year, and I didn’t feel like trying to do the math or redoing my spreadsheet, so everyone was just born in 1992 now. Like I said, gameplay-wise, it doesn’t matter.

Players have ‘attributes’, each scaled 0-99. Depending on how high they score out of 99 on those attributes is how you create their OVR. Which means I manually controlled things like how good players’ puck control, strength, endurance, offensive awareness, passing, wrist shot power, wrist shot accuracy, etc. are. For someone who watched a lot of NWHL this season, I based most of this on a combination of eye test and statistics. It’s not perfect, but players are often known for specific things- for example Thunstrom is known for her speed, Vlasic for faceoffs, and so on, which helps.

At first I made my scale 0-65. Then, the morning of the Final, I changed my mind and went back and tweaked everyone so that my scale was 0-100 the way you’d expect it to look. My highest OVR for a player is 92.

Creating Teams

I did my best to make things look the way you’d expect them to as far as team colors and numbers. I had a lot of fun messing with jerseys. Initially everyone only had home jerseys, but I ended up going back and giving the teams full jersey sets, partially for fun and partially out of necessity (you’ll see why eventually). The logo options aren’t awesome, but I think I did a decent job.

The trickiest part of the whole process was to test things out. I played each team against each other team to see if the matchups looked about the way we would expect them to based on the 2019-20 season. For example, if I had the Whale play the Pride and they absolutely slaughtered them, I knew I needed to tweak some things (no offense to the Whale).

This was around when I realized that the success of this endeavor also rested partially on how good I am at playing the game. I’m somewhere between a semi-pro and pro-level player (according to the settings, I am by no means a Professional NHL19 Player). If I play on semi-pro, I have to be careful not to overpower the AI and end up with a completely ridiculous outcome.

This is one of the reasons I switched teams by period in the Final — to give myself and the AI a sort-of equal shot at winning the thing. But ultimately, as long as the teams seemed to stack up the way I expected them to, I knew I was on the right track in terms of player ratings. Also, if the lines looked about the way they did in real life, it means that I did the OVR calculations correctly. It was very satisfying to look at the Whale lineup and see the lines looked exactly the way they looked during the last weekend of the Whale’s season, I won’t lie.

I’m planning on playing more games, and soon — so keep an eye out, stay home, and wash your hands!