Goldipucks and the Three Skaters: Duchene, Lee, & McAvoy
After a week's delay due to my 2019-20 Fearless Forecasts column, “Goldipucks and the Three Skaters” is back. For first time readers, or those needing a refresher, this column is a play on the Goldilocks and the Three Bears story, except instead of there being three bowls of porridge I’m covering three skaters and declaring one too hot (i.e., doing unsustainably better than he should), another too cold (i.e., doing unsustainably worse), and a third “just right” (i.e., producing where he should be). I also assign each a rating of 1-10, indicating just how hot (rated 7-10, where 10 is the most unsustainably hot), or how cold (rated 1-4, where 1 is the most unsustainably cold), or how “just right” (rated 4-7, where 5.5 is the most “just right”) he is.
This week I’m covering skaters who signed deals in the offseason: Charlie McAvoy, Anders Lee, and Matt Duchene. Can you deduce which was too hot, too cold, and “just right” for 2018-19? With the stakes being especially high in cap leagues, lock in your guesses and keep reading to see if you’re correct!
McAvoy followed up 32 points in 63 games as a rookie with 28 in 53 contests to stay right above the point per every other game threshold for his brief career. Looking more closely at his numbers, McAvoy saw his ice time rise each quarter last season, culminating in an average of 22:55 per game over his last 19 games (above his 22:10 average for the season), of which 1:44 was man-advantage time (above his 1:35 average for the season). Yet he managed just ten points in those games, meaning that even as he took the ice more his production didn’t really increase. It sure sounds like his scoring should stay about the same for 2019-20, and indeed that’s what the numbers and comparables seem to suggest.
Many have speculated that the presence of Torey Krug is holding McAvoy back; however, due to Krug missing time last season McAvoy was able to dress for 13 contests in which Krug didn’t play. In those games, McAvoy tallied eight points, which was clearly above his average. But he had zero PPPts, this despite seeing more than 2:00 of PP Time in six of the contests. Moreover, McAvoy’s low SOG rate (cumulatively 1.33 per game over his first two seasons) didn’t receive a boost, as he had 18 shots over those 13 games, for a nearly identical rate of 1.38 per game.
While on the subject of McAvoy’s SOG rate, it not only narrows the field in terms of player comparables but, arguably, more importantly, enables those comparables to perhaps take on more meaning. It turns out since 1990-91 there’ve been only three other d-men who, like McAvoy amassed at least 60 points in their first two seasons by age 21 while at the same time cumulatively averaging fewer than 1.5 SOG per game during those campaigns: Tyler Myers, Hampus Lindholm, and Tomas Kaberle. Myers made a big splash with 48 and 37 points in his first two seasons, but then saw his offensive production drop and went seven seasons before posting 30+ points again in 2017-18. Lindholm started with 30 and 34 points in his first two campaigns and, although still only 25, since then has yet to even match the 34. Kaberle topped 40 points in three of his first six seasons; but it wasn’t until after that time when we saw his full potential on display, with him scoring 53-67 points in his next three seasons, although then after that he bested 45 only twice more. What does this mean in terms of McAvoy? Maybe nothing, or it may suggest that if he ultimately becomes a fantasy success it might not be for quite a while, and it’s still very much an if, as shown by Lindholm and Myers.
Looking at McAvoy’s metrics, his 5×5 team shooting percentages were 8.91 as a rookie and 10.02 as a sophomore. Those are strong numbers, although perhaps influenced as much – if not more so – by his team as by him individually. His IPPs were 40.5% and 38.9%, which are quite decent for a defenseman, and show he’s involved in plays that lead to goals. His OZ% has been right at 55% both seasons, which yet again is somewhat a function of his team but does show he is being sheltered. As for his assists, he’s exactly equal in terms of primary versus secondary over his career, which is not technically a negative but also suggests he might be getting some lucky second assists. Overall, these numbers show McAvoy’s outputs are perhaps being bolstered by his team and that he’s certainly not been unlucky to date.
Perhaps, as many have presupposed, McAvoy will become a dominant fantasy d-man, especially if Torey Krug leaves Boston as a UFA in 2020; however, as we saw with McAvoy’s player comparables, that might not end up being the case at all or at least for a while, plus McAvoy didn’t exactly shine when Krug was out of the lineup for part of last season. Either way, for the time being, McAvoy’s production, is JUST RIGHT, as all his metrics don’t suggest good luck or bad luck and his ice times are likely to stay similar this season. I’d expect another season of point per every other game production, so he gets an exactly down the middle rating of 5.5.
Rather than testing the UFA waters, Islanders captain Lee opted to re-sign with the team this offseason for another seven years. But should we expect the 2017-18 version of Lee, who rose to 61 points to set a new career-high, or the 2016-17 and 2018-19 versions, which tallied just above 50 points? Based on the numbers, I’ll say neither – that is, he’ll be better than he was last season and 2016-17 but not quite as good as 2017-18.
Looking at 2017-18 and 2018-19 data and metrics for Lee, it’s remarkable how similar they were. He had 207 SOG in 2017-18 and 204 last season. His ice time was 17:15 per game in 2017-18 (versus 17:05 last season), of which 3:04 was on the PP (versus 2:51 last season). His 5×5 team shooting percentages were identical to the hundredth of a point, i.e., 9.47 in both the 2017-18 and 2018-19 seasons. And his IPPs were virtually identical at 59.8 in 2017-18 and 59.3 last season. And his OZ% was 50.1% last season versus just a slightly higher 52.7 in 2017-18.
So then, why the drop in scoring from 61 to 51 points? It was the team as a whole, which went from 264 goals scored to 228, for a drop of 15%, meaning that if Lee’s scoring were to have dropped by a comparable amount he’d have tallied 52 points. Okay, so why then is Lee not “Just Right?”
The big key is his shooting percentage had been 17.8% and 19.2% in the seasons since he became a top-six scorer, yet last season it fell to 13.7%. And that was despite the fact that his average shot distance shrank from 22.3 feet in 2016-17 and 21.7 feet in 2017-18 to 20.2 feet last season, suggesting he was more selective with his shooting and therefore, if anything, deserving of more goals via a higher personal shooting percentage And the difference is quite consequential, as had Lee’s personal shooting percentage been 18.5% (representing the combined rate of his personal shooting percentages from 2016-17 and 2017-18) instead of 13.7%, that would’ve meant nine more goals, putting him right back at 61 points.
But we can’t end the analysis there, as the Islanders PP sputtered, going from 58 PPGs in 2017-18 to a total of just 33 in 2018-19, for a drop of 43%. Yet Lee’s PPPts output fell from 22 to 16, for a drop of only 27%. Had his rate dropped by 43%, he’d have lost nine PPPts rather than six, which in turn would have meant him scoring 58 points instead of 61.
So there we have it – by the numbers Lee should have scored 58 points last season, not 51. And given how similar his data was for both last season and 2017-18, there’s little reason to suspect much change for those same numbers in 2019-20. Nor should we expect the Islanders to depart from the defensive focused philosophy that is Barry Trotz’s calling card. As such, since Lee’s production should’ve landed him at 58 points last season, he was TOO COLD; and he gets a rating of 2.0 to account for what should be a ten to 15 percent gain in-season to season points for him just in the normal course.
After spending the first eight seasons of his career with the Avs, Duchene played parts of the last two campaigns with Ottawa (107 points in 117 games) and then finished 2018-19 with the Blue Jackets, where he posted a lackluster 12 points in 23 games. Still, his combined totals for 2018-19 gave him 70 points in only 73 games, to hit the 70 point threshold for the first time since 2013-14 and second time overall. With that output last season and talk of him centering the top line in Nashville, where Duchene signed an offseason seven-year, $56M deal, he has to be bound for at least similar success? Not from where I sit, as despite his red hot start the data suggests he’s in line for lower scoring versus 2018-19.
Duchene’s 70 points in 73 games, for a point-per-game rate of 0.96 last season, was his highest by far since 2013-14, when he had 71 points in 70 games. Between then, Duchene’s cumulative scoring rate was 0.67, based on rates of 0.67, 0.77, 0.53, and 0.72 per season. Suddenly he rises to 0.96 per game when he’s playing for a new deal? Radar……going off. And the concern is not allayed by looking to his 2013-14 total of 70 points in 71 games because, lo and behold, he was playing for a new contract that season as well. So we have a player who scored nearly a point per game in a mere two of the last six campaigns (but otherwise at a 0.67 per game rate) and those two campaigns just happened to be right on the cusp of getting a new deal? Hmmm…….
But readers of this column know that my analysis would never stop there. We have to look at the numbers as well since perhaps there’s an explanation for why his scoring was so high in these two seasons and/or much lower in the other four. Looking at 2018-19, Duchene’s SOG rate per game was the lowest since his rookie season, and his personal shooting percentage his highest at 18.0% after being only 12.4% for his career prior to 2018-19, meaning he’d have scored ten fewer goals had he shot at the same percentage as he had for his career to date. There’s no explanation in his ice times either, as his PP Time was in line with his career average and his total TOI was only 20 to 30 seconds more per game than it had been dating back to, and including his 2013-14 other 70 point campaign. Moreover, although he only had 14 PPPts last season, that was the highest per game rate for him since…..you guessed it…..his 2013-14 other 70 point season. Digging deeper, last season he did have his highest ever percentage of time spent on the ice during his team’s PP Time, but it was not so much higher as to explain a marked increase in PP scoring.
What about his luck metrics? His IPP had averaged 73.9% in the four seasons since he last tallied 70 points, yet was 68.6% last season, meaning he probably lost close to a handful of points he otherwise would’ve received. Yet Duchene’s team shooting percentage was 10.76, after not being above 8.64 in the prior three seasons, likely due to his inflated personal shooting percentage. There’s also his OZ% of 57.5% from last season, marking only the second time it was above 50.06% since 2012-13. So although Duchene’s IPP was very low, his other luck metrics were higher than normal, making it so the points he may have lost via his lower IPP were either offset or even fewer in number than what he gained via his luckier than normal team shooting percentage and OZ%.
There’s also the reality that Duchene is now on the Predators, who, although more up-tempo than the squads coached for many years by Barry Trotz, still have not had a 65 point scorer this decade. Yes, Filip Forsberg did have 64 points in 67 games in 2017-18 and Viktor Arvidsson managed to tally 48 points in 58 games last season; however, those are the only two instances of players averaging 0.82 points per game or better since 2010-11. That’s a lot of seasons worth of data working against Duchene, to go along with what we already pointed out above.
All things considered, Duchene’s 2018-19 was TOO HOT. His production was outlying based on his prior results, the totality of his luck metrics, and given the timing of his contract yet. Moreover, although he’s going to a new team, it’s a squad that doesn’t breed high scorers. Even factoring in his hot start and high OZ% for 2019-20 so far, If he plays 80+ games I’d look for Duchene to score about 65 points, which is a far cry from his near point-per-game output from the 2018-19 campaign. Accordingly, he gets a rating of 9.25.
Mailbag questions needed
I’m always in need of questions to answer in my monthly mailbag column. There are two ways for you to get your questions to me – (1) email them to [email protected] with “Roos Mailbag” in the subject line, or (2) send me a private message at the DobberHockey Forums with your question (my username is “rizzeedizzee”.
No data at this moment.