In a hockey market as large and passionate as Toronto you're bound to get some pretty heated disagreements over how useful certain players are to the team. Opinions on Cody Franson have diverged widely for years, as some argue that he makes too many mistakes and doesn't skate well enough to hold down a top 4 spot, while others have been saying for years that he deserves a bigger role. Another Leaf who's been the subject of fierce debate is Tyler Bozak. While Brian Burke frequently said that the Leafs were playing Bozak over his head and needed to acquire a top line centre to ease the burden on him, he's been firmly entrenched as the first line pivot on the team for several years now.

Since the start of the 2013-14 NHL season, Bozak has put up points at a much higher pace than he did earlier in his career. This, too, has been the subject of violent argument. On one side are those who say that young Bozak is finally coming into his own, a late free agent signing out of the NCAA who needed time to adjust to the NHL game. On the other side are those who believe that Bozak is a mediocre player who has had his numbers boosted by a run of good luck and a lot of time playing with Phil Kessel (perhaps that's also a run of good luck). There are also those who fall somewhere in the middle, but they're boring, so we ignore them.

That leads to an obvious question - how many points is Tyler Bozak going to score this season? Is his increased scoring part of a new long-term trend, or is it a temporary blip that we should expect to recede? I tend to believe that in most circumstances, the best way to guess how a player will perform in the future is how he performed in the past. We might want to make adjustments for where the player fits along the aging curve, but generally the past is a pretty good indicator of the future. So I'm going to use Tyler Bozak's past scoring rate to try and guess where he might end up at the end of this season.

A Brief Diversion

But first, I just want to demonstrate why it's not usually a good idea to assume that because a player is performing well right now, he'll keep performing that way in the future. Here's Tyler Bozak's assist rate at even strength over the four seasons prior to this one:

Year A/60
2010-11 0.58
2011-12 0.96
2012-13 0.9
2013-14 1.68

Well that's a pretty interesting jump in his assist rate last season. A lot of people were convinced it was evidence that Bozak had finally come into his own as a playmaker. What does that chart look like if we add this year in?

Year A/60
2010-11 0.58
2011-12 0.96
2012-13 0.9
2013-14 1.68
2014-15 0.95

Yep, his assist rate this year is right back in line with the long-term trend.

Here's why I bring that up. Let's look at the same stretch of time, but this time we'll use points on the power-play.

Year PP P/60
2010-11 2.78
2011-12 3.59
2012-13 3.66
2013-14 3.11
2014-15 7.27

You can see that there's something going on here that's out of the ordinary. Now, maybe you think that Bozak has suddenly become an exceptionally good player on the powerplay. Maybe he has. But I think there's good reason to be very skeptical that that's the case. His point scoring rate while shorthanded has also seen huge boost this season. It's worth nothing that his goal scoring rate at even strength is actually lower than normal, so we could expect that to tick up a bit, though not nearly enough to make up for the expected drop in special teams scoring.

Projecting Bozak

So I've established why I think we should be skeptical of a player whose scoring in some situation or other has suddenly seen a massive spike. If that's the case, what should we actually expect from Bozak by the end of this season? We need to be careful not to fall into the Gambler's Fallacy here and assume that since he's been on a hot streak, a cold streak will necessarily follow. 23 games have already played, and those numbers are baked in. What I'm going to do is project how he'll do in the remaining 59 games based on his scoring rates so far in his career.

Here's how I've made my estimations: I've taken his previous goal and assist rates and multiplied them by the amount of ice time he'll get in the remaining 59 games if he keeps getting the same amount of ice time as he has so far this season. I've split these stats up into 5v5, powerplay, and shorthanded ice time to account for different rates of scoring in different situations. Maybe his ice time will change, but that's impossible for me to predict. It's also possible that he'll get injured, so my calculations based on 82 games could end up being wrong on that front.

I've provided projections for three scenarios here, though they turn out to be very similar. The first projection is based on Bozak's entire NHL career. The second projection is based on the past 3 seasons plus the beginning of 2014-15. As you can see in the tables I posted above, Bozak seems to have had a jump in his stats three years ago that's remained relatively consistent since. My final projection is based on the past 2 seasons plus the beginning of 2014-15. I think that's too small a window, but I've done it just in case anyone believes we need to look only at more recent data.

So, here are three different scoring projections for where Bozak will end up this season:

Years G A P
All 24 28 52
3+ 25 30 55
2+ 26 31 57

All three scenarios see Bozak setting career highs in goals and points. For the reason I stated above, I think the 3+ scenario is probably the best one to go with, though the differences here are pretty minor.

Is 55 points enough for a first line centre who plays with Phil Kessel and James van Riemsdyk? That works out to 0.67 points per game. Last year, among centres who played at least 60 games, that would rank 33rd, a bit outside of the bounds of what a 1st line centre does, especially given who Bozak plays with. But, of course, this projection could be all wrong. We'll have to revisit it at the end of the season and see if it did a good job of predicting how well Bozak will score.

[NOTE: The final paragraph originally had an arithmetic error that slightly understated Bozak's projected points per game. The post has been updated accordingly.]

How many points will Tyler Bozak have at the end of this season?

Less than 5043
50-59300
60-69322
70+68