Is Ty Smith Worse than Last Year, and Should the Devils Roster Him Anyway?

Every year, there’s the story about a guy that fights his way onto the roster that we weren’t watching for. Whether it’s Lee Stempniak, Brian Gibbons, Blake Coleman, Jesper Bratt, or someone else; there’s very frequently a surprise contender. This year, Matt Tennyson is making a case, for instance. However, more conspicuous this year, is someone who appears to be doing somewhat the opposite — Ty Smith’s performance seems to have moved him onto the roster bubble in my opinon.

Most (myself included) had Smith penciled in as the 6th defender this season since he tore up the WHL this season and almost made the NHL squad last year. It seems to me that his performance this pre-season, rather than cement his spot, has moved into a position more precarious than the onset of camp. Is this warranted?

Let’s start off with the pre-season numbers. Be aware, that these numbers ACTUALLY mean next to nothing. But given it’s likely the only time many Devils fans have seen Ty Smith play, I feel it may point somewhat to perception. According to NaturalStatTrick, he had no 5v5 points in either season, and 2 PP points both seasons (played ~40 minutes last year vs. ~60 minutes this year). He did take 6 shot attempts last year vs only 2 this season so we saw a little more offense. His CF%Rel went from -1.5 down to -4.7 this year, but his SCF%Rel went up from -9.9 to -5.9 (mostly due to defensive improvement). A big thing that I think people have noticed is the giveaways. After having only one 5v5 giveaway last pre-season, he’s accumulated seven (7) this pre-season — that’s the 2nd most among all NHL skaters so far. In particular, these giveaways have been in very dangerous areas — all in either the defensive zone or transition. NST has game reports with individual shot maps like this, and I spliced them together to show a map of Ty Smith’s giveaways in the 3 games so far on one map.

Now, his flaws were not limited just to these turnovers. Hynes gave a pretty lengthy list of issues Smith should be reviewing and has tasked defensive coach Alain Nasreddine to specifically go over these items with him. But it goes to a the larger point that it has not seemed as though Smith’s 200 foot game is anywhere near complete. I took a look at how he performed last season as just a bit of a fact check on my own memory. These were his stats last year vs. this year.

In the already muddy arena of preseason analytics, this is the corpse-filled marsh from Lord of the Rings. All I’ll say here is that it seems that he was allowing high-danger chances at a higher rate last year, but opponents converted on them at a crazy rate so far this year (4/9 vs 2/11 last year)— making his mistakes more apparent. Take that as you will, but I think some combination of “he wasn’t as good as he looked last year” and “bad luck” is true. I think a little bad luck has exposed the holes in his game that were present last year.

As I said a couple times, you can’t get much from pre-season stats. Everything up until now has been focused on what was fueling my — and perhaps many others’ — perceptions of him both Septembers. Now let’s look at his actual metrics from his WHL time. He went from 73 points in 69 games (1.06 P/GP) in his draft year to 69 points in 57 games (1.21 P/GP) this past season leading me to believe that he had officially outgrown his WHL surroundings. This was likely biased by my preconception that he had really looked the part in his pre-season games in NJ. When you plug further into the numbers, it’s a less clear picture.

If you look at his total point production, he went from a 74.9 points per 82 games to 97.8. That appears to be a fairly significant increase. But if you look at only his even-strength primary point rate, it went from 28.5 points per 82 games, to 30.21 — obviously an increase, though not nearly as noticeable or impressive. This is to say, his production increase is somewhat inflated by additional secondary assists, especially those on the powerplay.

For one more sanity check, I asked Byron Bader, who’s been sharing some really good prospect viz, for a consult on Ty Smith. He provided me with this image which shows Smith’s NHLe progression, in addition to the probability that he would become an NHLer/Star now versus at the draft.

As you can see, Smith has improved very slightly in the past season. His NHLe went up by just 4 points. His probability of becoming an NHLer has also increased by 12% though, which is likely a positive. I checked Manny Perry’s prospect model for confirmation and in his model Smith went from 13.1 to 13.9 NHLe, which is a comparably unimpressive improvement. He also had the probability that he’d make the NHL increase from 63% to 82%, but his projected WAR/82 went down from 0.21 to 0.13 — the fact that his plus-minus went from +49 (on a +42 team) to +9 (on a +45 team) likely hurt the assessment of his defensive skill.

Neither of these models separated by type of point, so the increased powerplay and secondary assists should bring down even these fairly conservative estimations of his growth. Which brings me to my final questions for you all.

1. Is Ty Smith better, worse, or the same as he was last year?
2. Is Ty Smith currently, or likely to become over the course of the year, one of the 6 best defencemen on this team?
3. Would playing in the NHL be the best decision for Ty Smith’s development?

I’ve attempted to set the stage as well as possible on these points, and I’ll let you offer your own opinions before interjecting mine, but I would like to point out an interesting possible combination of opinions for you to consider. Could the answer for #2 be “no,” but the answer for #3 be “yes”? If so, what should you do in that scenario? He’s not ready, but he didn’t seem to benefit from another season in the WHL, so maybe just get him going and see if he develops on the spot?

These are all tough questions and I look forward to hearing your opinions. Thanks as always for reading, and leave your thoughts in the comments section below!