The new season for the Lehigh Valley Phantoms is upon us, and despite the disappointing opener that we’re not going to dwell on here, it looks like it should be a good season and a fun team.
It’s also a team worth watching for those who like to keep an eye on the prospects. Because there are a lot of them! We have 11 skaters on the roster right now who we can still classify as prospects (though maybe you want to call them ten; the 11 includes Philippe Myers, who maybe you wouldn’t call a prospect anymore, and who may not be with the team for that long). But all the same, there are a lot of prospects to keep track of, and this season, there will be more and more in depths ways to keep track of them than ever.
If you love stats, you’re in luck, because we’ll have a lot of them at our disposal this season. And we’re here with a rundown of all that we’ll have publicly available.
First up, Brad’s back with a new and improved phancystats.com, and he’s got a breakdown of what’s going on there.
It’s year three of my Phantoms tracking project, and I’m excited to get back into this again. If you’ve used or looked at the data before, you’ll know that I never finished last season. It’s a lot for one person to tackle, and I burned myself out by game 50. So, in the hopes of that not happening again, I’ve changed things up.
The goal was to be able to get more data out of less, and I think that has been achieved.
Deciding what to remove was pretty easy — anything that is accessible on another website, and anything that I and others weren’t really using much. This led to the removal of point totals, primary points, shots on goal, faceoff wins and losses, and blocked shots. Dropping these also meant dropping Game Score, which will be missed as a very useful way to not only compare players, but to also quickly see each individual’s best performance of the season.
The final thing that was taken out of the project was goaltender statistics. They’d be nice to have, but in the end I decided that they weren’t worth the added time that it’d take for now. We’re back to just skater data.
New additions to site include scoring chances, high-danger Corsi, shots on goal differential, and PDO. Technically these were all a part of the project last season, but they were kept off-site. When it comes to scoring chances I think that it’s important to know exactly what the numbers that you’ll be looking at represent, and since there have been a few changes since I first started this, let’s have a full update on the process.
The first step in determining the quality of a shot is making note of where the shot was taken, or deflected. Two years ago I was working off of the War-On-Ice danger zones, as I explained in the introduction post from back then. This season I have my own danger zones based on both WOI’s findings, and Emmanuel Perry’s which can be found on Corsica. In creating mine I had to make sure that it both makes sense from a statistical standpoint, and that I can “see” the two zones while watching the games.
What I settled on was that shots taken from the middle of the interior hash marks down to the goal line would be high-danger chances. Easy enough. The medium danger zone, however, is a little bit tougher to put into words. Connecting itself to the high danger zone, the medium danger zone extends outwards from the top of the goaltender’s crease to the faceoff dots on either side. It then curves upwards, extending just above the faceoff circles at its highest point.
Below is a visual of the two danger zones.
A shot taken from anywhere else on the ice surface is a shot coming from the low danger zone. Shots from the high danger zone are given a value of 3, shots from the medium danger zone are given a 2, and, well, the rest are given a 1, just as you predicted. These number values are eventually what is used to bin every shot into three categories.
The second step is to determine whether the shot was a rebound, or if the shot was taken off of the rush. Both tend to lead to more dangerous shots, so for shots that qualify as either, their original value based on location gets upped by one. With rush shots, there are judgement calls where I won’t add one if the player with the puck, for example, enters the offensive zone and immediately lobs it on the net. That’s not a dangerous attempt.
Finally, one of the biggest advantages with manually tracked shot data is the ability to account for pre-shot movement. Two specific types of passes lead to more goals than others — the royal road pass, and passes that come from behind the net. Shots that come off of passes from the behind the net are pretty self-explanatory, and a royal road pass is a pass that occurs no higher in the zone than the top of the faceoff circles, where the puck crosses over the middle of the ice either left to right or right to left. Both pass types add one value point to the shot.
After all of that, shots with a value of “1” are counted as low danger, shots with a value of “2” are counted as scoring chances, and shots with a value of “3” or higher are counted as high danger.
Lastly, everything on the site this season is from a 5-on-5 game state, except for penalties drawn and penalties taken, which are gathered from all situations.
That’s it for me, but I am really excited about having Maddie’s neutral zone data from the get-go this season. We should have a lot of really good data to explore.
Speaking of the neutral zone data (which you can keep up with here), there have been some tweaks to that as well, but the breakdown is pretty self explanatory. We’ll be looking at the number of times a player skates the puck into or out of the offensive zone, makes a breakout pass or a pass as an entry, and when they just blindly dump the puck in or out. The percentages for entries are percentage of total entries that are carries (skating the puck in), controlled (skating it in or completing a pass to a teammate), and which end up with possession (skates, passes, and dump-ins recovered). Exits are much the same—we have percentage which are carries, and then percentage which are controlled (skates and passes again). And all of this is at 5-on-5. There will be season totals provided (and then broken up between forwards and defensemen), and then single game numbers will be provided as well.
Finally, the AHL is providing something new for us this season. In the single game logs, they’ll be tracking shot locations (again, only proper shots, not attempts). You’ll get a graphic at the end of the game that looks something like this. You can look at the shots period by period, as well, and click on a particular shot to see who took it.
We should make a note here that these locations are manually tracked, and as with all stats of that nature, aren’t perfect. But it’s a neat tool to have, if nothing else.
And that’s about all we’ve got! We’re excited for the season to get going, and to be bringing more data for everybody to dig into. It’s good stuff, folks.
