Announcer: Yes, folks… Moderna Designs present the latest in kitchen luxury. The Moderna Wonder Major All Automatic Convenience Center-ette gives you all the time in the world to do the things you really want to do… An infrared freezer-oven complex that can make you a meal from packet to plate in 15 1/2 seconds. –Time Bandits
Sadlly, we have no Moderna Wonder Major All Automatic Convenience Center-ette that could chart passing. If we could bring back the glow puck, along with glowing players, along with glow-enhancing technology, we could put little chips in our televisions and monitors and track how every event happens on the ice. I think the technology for that is probably a little ways off, so we have to make do with what we have.
When Gabe posted his request for a spreadsheet of events for the Detroit-Chicago game, I didn't see it. I did see the results, and thought that someone had made a graph in Paint or some other program tracking each pass by hand. I didn't realize that it was the product of a graph made in Excel. Silly me, I decided to start tracking passes by hand myself. I started out in paint with a mouse, but then remembered I had gotten a tablet as a gift from a rich friend I went to college with. She wanted her folks to buy her a new laptop, so she 'lost' her old one and I 'found' it. I hadn't had need of a laptop in a while, so I forgot all about it. But then I thought: it would be the perfect(maybe) tool for tracking passes by hand. I sprung into action.
What you see above is the result of tracking every pass, dump-in, and clear out in the first five minutes of the Ottawa Senators-New York Rangers game. These are the Senators' charts. The top rink represents the original hand drawn tablet passes, drawn in Windows Journal with a Lenovo X60 ThinkPad, while the bottom rink is what I recreated in Paint(also using the Tablet). If these charts look like a mess of lines, well, here's a handy guide on how to read them.
- On every graph, the left end is the Senators’ end, and the right end is the Rangers’ end.
- On the top graph, the dot on one end of the line represents the origin of the pass. The other end is where the pass ends up.
- If there is an “X” at the end of the line on the top graph, that means it was either a missed pass, an intercepted pass, or a turnover. You’ll notice the Senators had quite a lot of turnovers in the neutral zone. Some of this is a function of Karlsson, Kuba, and Gonchar making ‘ambitious’ breakout passes, but some is a function of me tracking every clear out, too. Some of those are good plays and not necessarily a result of bad passing, but the Rangers’ hard forecheck and them controlling the play.
- The “S” on the top graph represents a one-timer or a shot right after the pass. If a pass leads to a man breaking and getting a shot off-let’s say a breakaway, for instance- I did not include it. I wanted the charts to focus mainly on passing, and there are numerous shot charts out there.
- The top graph contains more of the dump-ins but the bottom graph does not, because (as you’ll see) this was an evolving process and I didn’t want to include dump-ins. That’s why you’ll notice that some of the dump-ins that wrapped around the net and have an “X” on the end(being stopped by Lundqvist) are not included on the bottom chart. I think this makes for a better chart.
- On the bottom graph, I didn’t originally include origins of passes. I thought it would be too clunky. I did include red circles as turnovers and green circles as shots. If there is a red circle that is in a faceoff dot position and not attached to a line, you can assume that it’s just a faceoff dot. Sometimes, a cigar is just a cigar.
- Last but not least, these are approximate passes. Obviously doing it by hand or by relative X/Y position is going to be approximate, but I tried my best to pinpoint the exact location of the passes.
Here are the Rangers' charts for the same timeframe:
The Rangers got off to a much better start than the Senators. The propensity of passing around the back of the net represents cycling, but you'll notice they weren't able to get too much sustained pressure and sometimes had difficulty gaining the zone. On dump-in passes(that is, soft dumps when there is a winger streaking) they had a couple of turnovers. The lines that begin just outside the Senators' zone and end in red dots in the Senators' zone are the result of good backchecking pressure in the neutral zone by the Senators' forwards and the defenseman beating the forechecker to the puck. The red dots behind the Senators' net are bad cycling by the Rangers forwards.
Okay, now it gets really interesting. The next five minutes the Senators really started to turn it on. They got a powerplay and generally controlled the pace of play- the Rangers had real trouble getting anything going
The lines in the Senators zone indicate a lot of back and forth passing by the Senators defense in low-pressure situations- right after changes, etc. You'll notice two or three stretch passes from the defense to the forward. One of them worked, one of them did not and the Rangers got a clear. Their powerplay movement was entirely around the outside, and their choice of shots looks strange. The four one-timers they tried were from relatively bad positions- close to the blue-line and on either side of the ice. They also could not, even on the powerplay, control the play behind the Rangers net. Now to the Rangers passing, and you'll see a marked difference in how few passes they achieved:
Again, in the top graph you'll notice more activity towards the top that go all the way into the Senators' zone as a turnover. These are clear-outs and I decided to exclude them from the cleaned up version. Excluding them, you'll see on the bottom graph that the Rangers had trouble making a breakout pass out of their zone, turning the puck over quite a bit in the neutral zone. There's a grand total of eleven or so passes for this time frame, and five of them were turnovers. I think there is a small bit of overlap here to the next Ranger charts, as I'm pretty sure that the only offensive zone to offensive zone pass that wasn't a turnover was Prospal's goal(in the northwest part of the chart). I could be mistaken, but I've tried to rectify that at the end with expanded tracking. On to the next, vastly different graphs.
What a change this is. Lots of turnovers by the Senators in this chart, and a lot of them coming in their zone or the neutral zone. Having played hockey(my resident 'I've watched a game before' quota), the turnovers are the result of hard forecheck at the half boards by the Rangers forwards. When they're playing well, they love to harass the team for the full length of the ice- a full court press, if you will. The Senators also had to kill off a penalty in this time frame, so you'll notice that they really didn't control the play or make smart passes in the offensive zone. The one one-timer they got was the result of a 4 on two rush. The other two times they made silly plays that resulted in turnovers.
The Rangers chart for the same timeframe is telling:
You'll notice that now they are controlling the play, with a lot of powerplay time, and thus have shifted their start from behind their net(or not even having possession) to the safer zone of above their faceoff circles. Though they had a couple of turnovers in the neutral zone, they were fairly successful all in all of gaining the zone and keeping it there. You'll also notice a difference in powerplay strategy. Although they also tried to work it up the side boards and get one timers from the point- the McCabe strategy- they also cycled a lot. And three minutes of this represents non-powerplay time, suggesting that they were controlling the pace and play and winning the battle of possession at even strength, too. One of these many cycles resulted in a goal by Prospal. The pass itself might be double counted, as it came right at the splitting mark for charting purposes. I tried to make sure this wouldn't happen again.
Having originally charted the remaining time the same way as I have above, I decided to go back and re-chart the remaining game also keeping track of four things- the passer, the intended target, the result, and the time. I wanted to satisfy some of the things Gabe originally asked for. This time, as with the third set of charts, I did not keep track of clear-outs or dump-ins or anything like that. I only wanted to keep track of intended passes. I broke the game down into a minute by minute basis:
I didn't bother cleaning these ones up in Paint- what you see here is exactly what I copied down in Windows Journal. The tablet pen works pretty darn well and is just as comfortable as writing with a real pen, so the messiness is actually my normal messiness. I can't blame my poor penmanship on the computer. But that's not what I wanted to show. This is both teams'('Sens' above the Senators and 'Rangers' above the Rangers) charts for the time frame of 4:00-3:00. I excluded the 5:00-4:00 because it actually contained less data than this one. But this is the main format I wanted to try to track it in.
- P is the passer, T is the target, R is the result, and Time is, well, the time as represented by the game clock. Assume for the ones that say ’48’ and ’27’ and so on that it represents 3:48 and 3:27.
- If the result is blank, that means it was just a completed pass. Otherwise, for example, pass 4 on the Senators’ chart has an “int 5” as the result, meaning Dan Girardi intercepted the pass. Since this is my first time, please excuse the errors like the lack of a singular format for “5 int” or int 5″ or whatever is written in the result column for Senators pass #2. I honestly have no idea what I wrote in the result column for the second Senators’ pass, but thankfully with the time and the players written down clearly, I can go back and look and see what it is. I also have it mapped as a turnover, so we’ll assume that it is that.
- I had to clear more space for the eleventh pass, so I decided to switch formats for the last three minutes. More on that in a second. Also, the MP in the eleventh Senators result column was a missed pass.
As you can see, this did not have a whole lot of data and I thought I could track it a little better with better use of space on the right side. So, now the really good ones(I think), the charts I'm really proud of:
This is the raw data- raw meaning exactly the way it looked in my tablet- for the last three minutes of the first period. The Senators controlled the play and had another power play but they still didn't work the puck low at all in the Rangers' zone. A couple more notes because I like bullet points.
- I assigned blame in this chart for missed passes. On the eleventh and tenth Senators’ passes, I honestly could not tell who Marek Svatos and Erik Karlsson were trying to pass to. And being a Ranger fan, I have experience with this from the old days of trying to figure out what Alex Kovalev was thinking. I couldn’t do it. So for the eleventh Senator pass I put a MPP which stands for “Missed Pass-Passer” meaning that the passer is to blame for that one. This is highly subjective so I would appreciate some feedback as to whether or not this should be kept in for future editions.
- On the tenth Senator pass I put “19/5 i” which is meaning both Ruslan Fedotenko and Dan Girardi should receive credit for blocking or disrupting the pass. If you look at the map you’ll notice two “X”s on the line. The first is where Fedotenko tipped it, and the second is where Dan Girardi stepped up an intercepted it.
- I counted a delayed shot and a tip dump-in in this chart. Delayed shot meaning the shooter could have one timed it but settled it down for a second before shooting. The ‘tip dump’ is what I’m calling it when the defenseman shoots the puck down the ice and the forward tips it in for a dump-in. I’m not sure I should be counting these, even though it’s sort of a pass. Again, feedback on this would be appreciated.
- The time frame for this is the last three minutes and in the time column I noted it every time a new minute happened, so I could write and track a bit faster.
I tried to keep the numbers of the passes easily visible on the map, and I think I might start putting them near the 'originating dot'(if you'll permit me to coin a term) to keep them viewable. As you might expect from watching the Sens, Karlsson's name sure does appear a lot in the passing column. Next time I do this, I would like to try and keep this data for the full sixty minutes to see if there is any correlation between anything- I'm not that good with stats but maybe someone can find something in the numbers, if someone who is better at Excel wants to enter the data.
For the last three minutes I cleaned it up in Paint, too.
These are the Senators last three minutes with a little upgraded chart. I put black dots in the Paint versions to represent pass origination and I put red circles to represent a missed pass or a two-part interception(Fedotenko and Girardi from earlier, for example). I think this makes the cleaned up version a lot more interesting and helpful. In particular, you can see how Spezza and Karlsson, both on their off-wing, like to play catch along the side boards on the power play. And how the Senators tried little east-west passes at the blueline and kept turning it over. And again, they have no sustained cycling presence at all, except in their own end.
By contrast the Rangers didn't turn the puck over, but only because they didn't have the puck nearly at all for the last bit of the period. The main thing I notice here is that they have one pass in the cycling area, a Derek Stepan one-timer from a weird area.
Also, in this graph I put the number of the pass in, in the Paint version. This is probably the way I am going to chart passes from now on. I put the numbers right on the origination point and I think it makes for a very readable chart. I left them out of the Senators' charts because I wanted to get feedback if they seem redundant. I think it's the way to go, personally. It might get clunky if there is an abundance of passing(I want to chart a Red Wings game because I am a masochist), but I think it will work out. And it's really not all that time consuming.
So, if you've gotten this far, thank you for reading. If you come out to Tennessee some time, I'll buy you a beer. I've never done this before, but it was really fun and it illuminated certain things(notice how the Rangers almost exclusively moved the puck up the left side in the period) that I might not have noticed other wise, or been able to put my finger on.
I'm really excited about this because I think having a tablet cuts down on the time to make these things a bunch. The old argument about electric and hydrogen cars- if the price comes down, then people will drive it- also suits event mapping in hockey. If the time comes down, then more people will do it. Gabe said that the Hawks fan who did the original charts spent eighteen hours on it. That's crazy(and exactly what I would expect from hockey fans), but it'll be hard in the absence of new technology to chart events if it takes that long. As I got more comfortable and refined my methods, the mapping and tracking took a lot less time than it did when I started.
Gabe has also offered me a platform for a number of ideas that I have and I couldn't be more thankful to him, because there's so many smart people who read this site and contribute. Some of the ideas that I would like to do(many of them tablet based, because if you have a player like Jagr, you get the puck to Jagr, so to speak) are tracking scrums along the boards to see who comes out with the puck most often(I would bet Ryan Callahan is better at this board play than Marian Gaborik, for instance), tracking a player's skating pattern over the course of the game(Duncan Keith and Nicklas Lidstrom are both great, but I think Keith's would be more active and cover more ice than Lidstrom's), and tracking dump-in patterns and clear-outs in a separate chart to see who is doing what and if it correlates with on-ice success as much as it intuitively seems.
So thank you so much for reading again and please let me know what you think. And as a feeler, if anyone with spreadsheet skills wants to volunteer to help me with the next set of charts, please email me or post a comment. My email is [geoffbok] [@] [gmail.com]
Cheers,
– Geoff
