Hey look, hockey is coming. With training camp starting this weekend, and actual real NHL hockey starting later this month, I figured it might be time to finally post my goal predictions for the 2012-13 season. So for this post, I did an obscene amount of research, spent an insane amount of time crunching numbers, and poured over data from many years relating to every player on this team. The problem is that all this time was spent back in September. So the numbers sat, and sat, and sat, and sat, and sat, until now when there really isn’t a 2012-13 season anymore, and goal projections for an 82 game season make no sense. Nor does a lot of the logic behind how I came to those conclusions. Sigh.
In spite of this, I’m still going to present my full season goal projections. I don’t care if it doesn’t entirely make sense to provide a prediction for an 82 game season. I don’t care if a 48 game season isn’t a big enough sample size for shooting percentages to balance out the way they should. I don’t care if the shortened schedule will likely play a big role in how much ice time players get (condensed schedule likely equals more balanced ice time plus more injuries). The reality is a prediction for how many goals RJ Umberger will score in a 48 game season means nothing to anyone. If I predict he’ll score 23 goals in an 82 game season, there is a context for how good (or bad) that is. If I tell you I think RJ will score 13 goals this year, that means nothing (without doing some math).
So onto the predictions/projections. The roster turnover in the off-season made this a lot more difficult than it has been in years past. Historically, players tend to produce shots on goal at a fairly standard rate over the course of their careers. The biggest variances in terms of goal scoring come in shooting percentages and time on ice. So if we can estimate a player’s time on ice at even strength and at 5v4, and use the league average shooting percentage, we can pretty easily come up with some projections. Only the elite of the elite have ever shown the ability to score on an above average number of their shots on goal. The current incarnation of the CBJ don’t have any of those players. The problem with this roster was figuring out the time on ice. I used a range of possibilities, from the worst probable outcome to the best probable outcome to determine their range. To use RJ as an example again, the worst possible outcome had him miss a number of games due to injury, end up playing mostly 3rd line even strength minutes, and getting little powerplay time. The best case for him involved 1st line ES minutes, top unit PP time, and no games missed to injury. After determining the ranges, I put in what I thought the CBJ roster would be for the coming year, who I thought would play on what line, what PP unit, what defensive pairing, etc. I used their average shot rate over the last three seasons, unless they don’t have three seasons in the league, then I used an average of the three most comparable seasons from other players around the league over the last few years (unfortunately I did not write them down, but I know that one of the comps for Johansen this year is Seguin last year). The following chart has my low goals estimate, high goals estimate, my ES goal prediction, PP goal prediction, and a pro-rated goal total for a 48 game season.
*Yes, I know Ryan Murray is injured and won’t be with the CBJ this season. But I did the work to figure out his comparable players (Drew Doughty, Victor Hedman, and Erik Gudbranson) and how their shot rates would average out over the time I expected Murray to play pre-season. So I’m keeping him in.
So what does this all mean? First, that the Jackets are definitely going the scoring by committee route. Second, that one of Atkinson or Johansen will have to have a very large breakout year for the team to jump way up in scoring. And finally, that the scoring by committee approach should work (provided the team stays healthy). My prediction puts the Jackets at 153 even strength goals over an 82 game season. This is a jump of 21 goals over last year. 153ESG’s would have ranked the CBJ 14th in the NHL (sandwiched between Buffalo and Washington), and 6th in the Western Conference. This compared to last year when they put up a measly 132 for a 27th place finish.
As for the powerplay, my projection puts them at a total of 43 5v4 goals over an 82 game season. This one is a little tougher to quantify. 43PPG’s would rank the Jackets 11th last season, although they did finish last season with 42 5v4 goals (42 5v4 goals is league average). However, the sum of the 5v4 goals isn’t really a great indicator of powerplay quality. The Jackets finished the season with over 424 minutes of time spent at 5v4 (second in the league), and almost 40 minutes more than the 3rd place team (3rd place Montreal is closer to 8th than they are to 2nd). I highly doubt the Blue Jackets will spend that much time on the powerplay this coming season, as they have traditionally been around the middle of the pack, as were Todd Richards Minnesota teams. However, a decline in time on the PP should be outweighed by a full season of Jack Johnson, James Wisniewski, Cam Atkinson, and the addition of several other skilled players. Hopefully, Derek Dorsett will spend the more reasonable amount of zero minutes on the PP this year.
If you can add, you have now figured that I have the Jackets at 196 goals for this season. If you are familiar with team goal totals, you also know that this is not a good number. That’s because there are a lot of goals missing, and historically those goals have: a) not mattered as much as people think; and b) come in such a limited sample that they are not predictable. On average, we can expect the CBJ to score another 26 goals. This includes 3 5v3 goals, 7 4v4 goals, 1 4v3 goals, 5 4v5 goals, 8 empty net goals, and 1 penalty shot goal. You may be thinking that certain things, like winning teams and empty nets goals, go hand in hand. Well they don’t (exactly). Last season, the last place Jackets finished in the top 10 in the NHL in empty nets goals scored. You might thinkg that high scoring 5v4 powerplays should equate to high scoring 5v3 PP’s. Well the 28th ranked 5v4 team, Montreal, finished 1st in 5v3. Similarly, the Avalanche was ranked 24th at 5v5 and 3rd at 4v4. The Red Wings were 3rd 5v5 and 24th 4v4. Basically, there is no rhyme or reason to these goals. These situations just don’t happen enough to cull enough data to figure anything out.
If you add these three numbers, you know that I have the Jackets at 222 goals over an 82 games season. This would have them tied with the Rangers for 11th in the NHL last season, a significant jump from their 28th ranked 198 goal output. This would also (sadly) mark the franchise record for goals in a season. This realistic, yet optimistic outcome would see only RJ Umberger definitely top 20 goals, although those other 26 goals might be spread around in a way to get a number of other forwards (notably Cam) over the 20 goal hump. However, it would also see at least 9 forwards with double digit goals, with a solid chance that Johnson, Wisniewski, and Nikitin also join those ranks by picking up 2-3 of the random 26 goals.
So while Rick Nash’s 30 goals will be missed, the additional scoring bodies brought in will actually help raise the CBJ’s scoring. Last season saw 6 forwards top 10 goals, and only one defenseman do so. Wiz and Niki both were on double-digit goal pace last year, with injuries and Nikitin’s time in St. Louis dragging those numbers down. Injuries is the key part of this, and my predictions keep those in mind – giving a half season each to depth players Gillies, Russell, Moore and Erixon, and taking away that time from the more injury prone guys higher in the lineup. So barring devastating injuries to multiple players, the team should have the depth to push through (and this is without considering better scorers like JAM, Calvert, Kubalik, etc. in place of Gillies/Russell). Similarly, the defense has equal depth (although you can bump Murray out, Erixon/Moore in, and add Savard as an injury replacement, which will basically all be a wash scoring-wise).
Obviously this does not take defense or goal-tending into consideration. But this forward group is a much better set of two-way guys, the defense is much deeper than last year, and Bobrovksy-Mason can’t be as bad as the Sanford-Mason-York group was last year. If the team allows 1 less shot against per game, and the Bob-Mason-CMac group brings the collective save percentage from .903 to .910 (still below league average), the Jackets would drop their goals against total from 28th ranked 257GA to a 17th ranked 220 goals against. Those are both very modest and realistic outcomes. Combine 220 goals against with 222 goals for, and you have the the very first Columbus Blue Jackets season to result in a positive goal differential. Last season every team with a positive goal differential made the playoffs.